SYSTEM AND METHOD FOR
RECOMMENDATION IN APPRAISAL OF
MICROFINANCE LOANS
BACKGROUND
[0001] The invention relates generally to microfinance and, more particularly,
to an automated technique for loan appraisal in a financial institution offering microcredit.
[0002] Microfinance refers to the provision of small-scale financial services (for
example loans) and sustainably delivering such services to low-income consumers for their occupational needs. The low-income consumers may include farmers, fishermen, lower skilled laborers, owners of small shops or microenterprises, and so forth, in both urban and rural areas. Typically, these consumers are not served by the commercial banks, as they do not have steady and sufficient income. Further, they seldom have anything of value to provide as collateral. Thus, generally the micro-loans are provided to these consumers without any collateral. The benefits of such loans to the microfinance entities are that they are low on risk and high on returns. For the low-income consumers, the benefits are to increase the earning of the consumers through financing thereby easing or in some cases eradicating poverty. Because of the higher rate of returns and lower risk, many banks and financial institutions have ventured into this field.
[0003] As illustrated in FIG. 1, the process of microfinancing involves
recruiting local people who personally know the community and area, as agents, and educating the agents about the various financial services and products. Each agent then makes field trips and forms a group of people who can vouch for each other. Each group is then educated about the financial services, repayment procedures, group benefits and responsibilities, and so forth. The agent's supervisor or the branch manager then meets with these groups and verifies the depth of their knowledge. The members of the group are then approved or retrained based on their knowledge. One or more members of the approved group may then raise a request on behalf of a customer for various financial services such as a request for a loan. The branch manager manually reviews the request and sanctions or denies the request. Sanctioned loans are then disbursed through the agents to the consumer. The agent may then collect the periodic installments from the

consumer at periodic and regular intervals and record these periodic installments using
manual receipts. The data from these collection receipts are then manually entered or
transferred into the management information system (MIS) of the microfinance entity.
[0004] The information required to be processed while appraising a loan
application is very different from those required by the commercial banks because of the
different class of customers targeted by the microfinance institutions. The lack of steady
income or collateral makes the loan appraisal process very challenging for the branch
manager or any other loan appraisal manager. The income, expense, asset and liability
information provided by the customers are generally rough estimates. The manager
therefore uses their personal understanding of the socio-demographic status of the
customer and their personal understanding of how correct the estimates are, to appraise
the loan application. The manager may seek input from respective agents to develop an
understanding of the customer and the customer's income potential.
[0005] Thus, typically in the field of microfinance, the loan appraisal process is
carried out manually and is based on subjective inputs. The branch manager has to
invest a lot of time and effort with each loan application. Though many automated loan
recommendation techniques exist in the banking domain, such techniques do not exist or
are not configurable to work in the microfinance domain. The primary reason for the
lack of automated techniques for the loan appraisal process is that the kind of people who
are addressed by microfinance entities fall well below the poverty line and hence are
unable to produce any kind of security (collateral) against the money borrowed.
[0006] As the microfinance institutions are trying to scale up their operations to
cater to growing demands, the microfinance institutions either need to increase the workforce of agents for the loan appraisal process by multiple times or achieve operational efficiency through the use of technology. However, the microfinance entities find it difficult to hire and retain people. In other words, the field of microfinance is at a stage of rapid growth due to huge market potential but lacks appropriate technology to enable the rapid growth.
[0007] It is therefore desirable to develop an automated and efficient technique
for the loan appraisal process to assist the branch managers in making quick loan disbursement decisions so that the microfinance institutions can achieve greater reach and speed while minimizing the cost of operations.

BRIEF DESCRIPTION
[0008] hi accordance with aspects of the present technique, an adaptive loan
appraisal system for assisting loan appraisal personnel in making a loan disbursement decision is presented. The adaptive loan appraisal system includes a loan recommender module configured to review a current loan request based on input data and one or more associated parameters to provide a recommendation for a current loan request, and adaptively adjust the one or more associated parameters based on a final decision of the loan appraisal personnel regarding the current loan request.
[0009] In accordance with another aspect of the present technique, a device for
assisting loan appraisal personnel in making a loan disbursement decision is presented. The device includes an adaptive loan appraisal system, where the adaptive loan appraisal system includes an input module for accepting input from a loan applicant, a gross surplus module for computing a range of gross surplus income based on an income potential, an expenditure pattern, and occupation of the loan applicant, a net surplus module for determining a range of net surplus income based on the range of gross surplus income, previous loan details and corresponding periodic repayment installments, and a loan recommender module for reviewing a current loan request based on input data and one or more associated parameters to provide a recommendation for a current loan request, and adaptively adjusting the one or more associated parameters based on a final decision of the loan appraisal personnel regarding the current loan request.
[0010] In accordance with yet another aspect of the present technique, a method
for assisting loan appraisal personnel in making a loan disbursement decision is presented. The method includes accepting input data from a loan applicant, reviewing a current loan request based on the input data and one or more associated parameters to provide a recommendation for a current loan request, and adaptively adjusting the one or more associated parameters based on a final decision of the loan appraisal personnel regarding the current loan request.
[0011 ] In accordance with yet another aspect of the present technique, a method
for assisting loan appraisal personnel in making a loan disbursement decision is presented. The method includes accepting input data from a loan applicant via an input module, computing a range of gross surplus income based on an income potential, an expenditure pattern, and occupation of the applicant via a gross surplus module,

determining a range of net surplus income based on the range of gross surplus income, previous loans and corresponding periodic installments via a net surplus module, reviewing a current loan request based on the input data and associated parameters to provide a recommendation for the cxirrent loan request via a loan recommender module, adaptively adjusting a minimum value and a maximum value associated with a purpose of the current loan request via a loan recommender module, and adaptively adjusting the one or more associated parameters based on a final decision of the loan appraisal personnel regarding the current loan request via the loan recommender module.
DRAWINGS
[0012] These and other features, aspects, and advantages of the present
invention will become better understood when the following detailed description is read
with reference to the accompanying drawings in which like characters represent like
parts throughout the drawings, wherein:
[0013] FIG. 1 depicts a generic process followed by microfmance institutions to
provide financial services to consumers;
[0014] FIG. 2 is a schematic of an exemplary adaptive loan appraisal system, in
accordance with aspects of the present technique;
[0015] FIG. 3 is a schematic of an exemplary microfmance system including
the adaptive loan appraisal system of FIG. 2, in accordance with aspects of the present
technique;
[0016] FIG. 4 is a schematic of an exemplary device including the adaptive loan
appraisal system of FIG. 2, in accordance with aspects of the present technique;
[0017] FIG. 5 is a flowchart illustrating an exemplary automated loan appraisal
process, in accordance with aspects of the present technique;
[0018] FIG. 6 is a flowchart illustrating an exemplary method for determining a
range of gross surplus income of a loan applicant, in accordance with aspects of the
present technique;
[0019] FIG. 7 is a flowchart illustrating an exemplary method for determining a
range of net surplus income of the loan applicant, in accordance with aspects of the
present technique; and
[0020] FIG. 8 is a flowchart illustrating an exemplary method for reviewing a
current loan request, in accordance with aspects of the present technique.

DETAILED DESCRIPTION
[0021] The present techniques are generally directed to automating loan
appraisal processes in the area of microfmance, microlending and microinsurance
through an adaptive loan appraisal system. Such an automated technique and system for
enabling the automation of the loan appraisal process in the area of microfmance may be
useful in a variety of other contexts, such as commercial banking, institutionalize
banking, and so forth. Though the present discussion provides examples in context of
microfmance, one of ordinary skill in the art will readily appreciate that the application
of these techniques in other contexts is well within the scope of the invention.
[0022] Turning now to FIG. 2, a schematic diagram of an exemplary adaptive
loan appraisal system 10, in accordance with aspects of the present technique, is
illustrated. The adaptive loan appraisal system 10 may be configured to assist loan
appraisal personnel, such as a branch manager in a microfinance institute, in making a
loan disbursement decision. More particularly, the adaptive loan appraisal system 10
may be configured to review a current loan request from a loan applicant based on input
data provided by the loan applicant and one or more associated parameters to provide a
recommendation regarding the current loan request. Additionally, the adaptive loan
appraisal system 10 may be configured to adaptively adjust the one or more associated
parameters based on a final decision of the loan appraisal personnel, thereby facilitating
generation of more informed recommendations regarding loan requests.
[0023] In a presently contemplated configuration, the adaptive loan appraisal
system 10 may include an input module 12, a gross surplus module 14, a net surplus module 16 and a loan recommender module 18. The input module 12 may be configured to accept input data from the loan applicant. The input data may include income details, expenditure details, occupation of the loan applicant, current loan request requirement details, purpose for the current loan request, previous loan details, if any, or combinations thereof More particularly, in certain embodiments, the loan applicant may input one to three sources of expected income. These sources of income may be input as discrete values. Moreover, the sources of income may include fixed sources of income, variable sources of income, or both. Furthermore, the variable sources of income may include variable sources of income with seasonality or variable sources of income without seasonality or both. As used herein, the term "variable source of income with seasonality" is used to refer to a source of income that varies based on a season. Also, as

used herein, the term "variable source of income without seasonality" is used to refer to a source of income that does not vary based on a season. By way of example, the source of income of a farmer may be representative of a variable source of income with seasonality as the income may increase during the harvest season and diminish during a planting season. Further, the income of a petty businessman may be representative of a variable source of income without seasonality as the income is not dependent upon any season. It may be noted that if the source of income includes a variable source of income with seasonality, then income during a lean season and a peak season may be input by the loan applicant. In addition, months associated with the lean and peak periods may also be provided as input by the loan applicant.
[0024] Furthermore, for every source of income input by the loan applicant, the
input module 12 may also be configured to accept as input a corresponding source of expenditure. Here again, the sources of expenditure may be input as discrete values. The input module 12 may also be configured to accept as input details related to the occupation of the loan applicant. The occupation of the loan applicant may be entered as free text. By way of example, the occupation of the loan applicant may include farmer, barber, auto mechanic, petty businessman, potter, mason, blacksmith, and the like. The adaptive loan appraisal system 10 may be configured to store all the occupations in a data repository, such as data repository 20. The listing of occupations may be employed to aid in providing a recommendation regarding the current loan request, such as an approval or rejection of the current loan request. It may be noted that although the data repository 20 is shown in FIG. 2 as being detached from the adaptive loan appraisal system 10, in certain embodiments, the data repository 20 may be an integral part of the adaptive loan appraisal system 10.
[0025] The input module 12 may also be configured to accept details of
previous loans of the loan applicant, if any. The details of the previous loans may include the loan amount, tenure of the loan, and source of the loan. In addition, if available, the loan applicant may also provide the interest rate and the periodic repayment installments corresponding to the previous loans as input via the input module 12.
[0026] Moreover, the input module 12 may also be configured to accept the
purpose of the current loan request. The purpose of the current loan request may include the occupation listed by the loan applicant or certain actions associated with the occupation. It may be noted that the purpose for the current loan request may also be

different from the occupation of the loan applicant. For example, a loan applicant who is
a barber by occupation may apply for a loan to repay a previous loan borrowed at a high
interest. Each purpose input by one or more loan applicants may also be stored in the
data repository 20, for example. Furthermore, each purpose may be categorized into a
set of various sub-items. Also, a minimum currency value and a maximum currency
value may be assigned for each sub-item. In certain embodiments, the loan appraisal
persormel may assign the minimum and maximum currency values for the sub-items.
This information may be stored in the data repository 20. In accordance with aspects of
the present technique, this information may be stored in the form of a table. This table
may be referred to as Purpose Break Up Table. Subsequently, if a similar purpose for
another loan applicant arises, the adaptive loan appraisal system 10 may be configured to
automatically popup all the associated sub-items and the corresponding maximum and
minimum currency value requirements to accomplish the sub-items to aid the adaptive
loan appraisal system 10 in determining a recommendation regarding the current loan
request. Additionally, the input module 12 may also be configured to accept as input an
amount (expected loan amount) corresponding to the current loan request.
[0027] The exemplary adaptive loan appraisal system 10 may also be
configured to compute a range of gross surplus income of the loan applicant. More particularly, the gross surplus module 12 may be used to compute a range of the gross surplus income range of the loan applicant, in one embodiment. Specifically, the gross surplus module 12 may be configured to compute the gross surplus income range of the loan applicant based on the income details, the expenditure details and the occupation of the loan applicant. It may be noted that the gross surplus module 14 may be configured to determine an income potential of the loan applicant based on the income details provided by the loan applicant. In a similar fashion, an expenditure pattern of the loan applicant may be generated by the gross surplus module 14 based on the expenditure details provided by the loan applicant. It may be noted the gross surplus income range of the loan applicant may also be dependent upon demographics and lifestyle of the loan applicant. Additionally, the gross surplus income range of the loan applicant may also be dependent upon seasonality factors associated with the source of income and/or expenditure. The determination of the gross surplus income range of the loan applicant will be described in greater detail with reference to FIGs. 3-7.
[0028] Moreover, the adaptive loan appraisal system 10 may also be configured
to compute a range of net surplus income of the loan applicant. The net surplus module

16 may be used to compute the net surplus income range of the loan applicant, in one embodiment. More particularly, the net surplus module 16 may be configured to compute the net surplus income range of the loan applicant based on the gross surplus income computed by the gross surplus module 14, previous loans, if any, and periodic repayment installments associated with the previous loans of the loan applicant. The determination of the net surplus income range of the loan applicant will be described in greater detail with reference to FIGs. 3-7,
[0029] Additionally, the adaptive loan appraisal system 10 may also be
configured to provide a recommendation regarding the current loan request to the loan
appraisal personnel. The recommendation may include an approval of the current loan
request or a rejection of the current loan request. In certain embodiments, the
recommendation may also include holding the current loan request in abeyance for
fiirther review. In one embodiment, the loan recommender module 18 may be employed
to evaluate the current loan request and provide a recommendation regarding the current
loan request. More particularly, the loan recommender module 18 may be configured to
provide the recommendation based on the current loan request, the net surplus income
computed by the net surplus module 16 and one or more associated parameters.
[0030] hi accordance with further aspects of the present technique, the adaptive
loan appraisal system 10 may be configured to use one or more associated parameters to facilitate generation of the recommendation regarding the current loan request. In one embodiment, the associated parameters may include an acceptance threshold value Acceptance Threshold, an acceptance limit value Acceptance Limit, a threshold increase level value Threshold Increase Level, and an experience threshold value Experience Threshold. The Acceptance Threshold is representative of a confidence threshold that the adaptive loan appraisal system 10 maintains for an expected loan amount. In one embodiment, the expected loan amount may be representative of an amount corresponding to the current loan request. Moreover, in one embodiment, the Acceptance Threshold may have a value of about 0.6. However, this value may vary with the purpose of the current loan request. Furthermore, the Acceptance Limit may be used to define a number of loan requests that need to be approved before it is desirable to modify the Acceptance Threshold. Initially, the Acceptance Limit may have a value of 5. Moreover, the Threshold Increase Level may be used to determine a percentage increase in the Acceptance Threshold once the Acceptance Limit is reached. In one embodiment, the Threshold Increase Level may be set to a 10% increase. Additionally, the Experience

Threshold is used to refer to a desirable number of years of experience of the loan
applicant in a particular occupation. In one embodiment, the adaptive loan appraisal
system 10 may be configured to maintain only one value indicative of the Experience
Threshold. In certain embodiments, the Experience Threshold may have a value of 1
year. However, the adaptive loan appraisal system 10 may also be configured to
maintain different values of the Experience Threshold for different occupations. In one
embodiment, the associated parameters may be stored in a data repository 20, for
example. Alternatively, the associated parameters may be stored in a management
information system (for example MIS 36) in a central location (for example central
office 32) or at each branch location (for example branch office 34).
[0031] In accordance with aspects of the present technique, the loan
recommender module 18 may be configured to determine a periodic repayment
installment corresponding to the current loan request. Furthermore, the loan
recommender module 18 may also be configured to compute a confidence level for the
current loan request. As used herein, the term "confidence level" is used to refer to a
relative distance of an expected loan amount corresponding to the current loan request
from a maximum allowable value. Moreover, the loan recommender module 18 may
also be configured to determine repayment capability of the loan applicant. Specifically,
the loan recommender module 18 may be configured to determine the repayment
capability of the loan applicant based on the net surplus income range computed by the
net surplus module 16 and the periodic repayment installment corresponding to the
current loan request. Moreover, in accordance with further aspects of the present
technique, the loan recommender module 18 may also be configured to determine an
experience value of the loan applicant. As used herein, the term "experience value" is
representative of experience of the loan applicant in a general area associated with the
purpose of the current loan request provided by the loan applicant as input data.
[0032] Once the confidence level, the repayment capability and the experience
value of the loan applicant are determined, the loan recommender module 18 may be configured to review the current loan request based on the confidence level, the repayment capability, the experience value of the loan applicant and the associated parameters. As noted hereinabove, the loan recommender module 18 may review the current loan request to provide a recommendation to the loan appraisal persoimel, for instance. The recommendation of the loan recommender module 18 may include an approval of the current loan request or a rejection of the current loan request. Also, the

recommendation may include holding the current loan in abeyance for fiirther review.
Furthermore, the loan recommender module 18 may be configured to communicate the
recommendation to the loan appraisal personnel. The determination of the
recommendation will be described in greater detail with reference to FIGs. 3-7.
[0033] In addition, the adaptive loan appraisal system 10 may also be
configured to accept a final decision of the loan appraisal personnel. The loan appraisal personnel may make a final decision based on the recommendation provided by the adaptive loan appraisal system 10 and awareness of field information of the loan appraisal personnel. As used herein, the term "field information" is used to refer to loan appraisal personnel's understanding about the purpose for which the current loan is sought, the occupation of the loan applicant, cash inflow and outflow of the loan applicant, peak and lean periods associated with the occupation of the loan applicant, the socio-demographic status of the loan applicant and loan appraisal personnel's understanding of the accuracy of the estimates provided by the loan applicant, or combinations thereof.
[0034] In accordance with exemplary aspects of the present technique, the
adaptive loan appraisal system 10 may be further configured to adaptively adjust the one or more associated parameters based on the final decision of the loan appraisal personnel. The adaptive loan appraisal system 10 and more specifically the loan recommender module 18 may be configured to adaptively adjust the associated parameters, such as the Acceptance Threshold, the Acceptance Limit, the Threshold Increase Level and the Experience Threshold, based on the final decisions (approval of the several loan requests) of the loan appraisal personnel. By way of example, in one embodiment the value of the Acceptance Threshold is set to 0.6. This value of the Acceptance Threshold may be adaptively adjusted if the adaptive loan appraisal system 10 receives at least 5 acceptances of loan requests {Acceptance Limit). Subsequently, the Acceptance Threshold may be increased by 10% (Threshold Increase Level). In a similar fashion, the values of the Acceptance Limit and the Threshold Increase Level may also be adaptively adjusted by the adaptive loan appraisal system 10. Consequently, the adaptive loan appraisal system 10 is adapted to "learn" with each final decision of the loan appraisal personnel, thereby facilitating generation of subsequent recommendations that are "more informed" recommendations. Specifically, the adaptive loan appraisal system 10 is configured to adjust the associated parameters based on the final decision of the loan appraisal personnel. In one embodiment, the updated associated parameters

may be stored in a data repository 20, for example. Alternatively, the updated associated parameters may be stored in a management information system (for example MIS 36) in a central location (for example central office 32) or at each branch location (for example branch office 34).
[0035] Referring now to FIG. 3, a schematic diagram of a microfinance system
30 including the exemplary adaptive loan appraisal system 10 of FIG. 2, in accordance with aspects of the present technique, is illustrated. The microfinance system 30 includes a central microfinance institute 32 that provides financial services to consumers through its various branch or regional offices 34. All of the information of the central microfinance institute 32 and the branch offices 34 are maintained in their respective information management systems (MIS) 36. The MIS 36 may therefore store information such as consumer names, their loan details, their payment history, request for new loans, loan approval or disapproval details, past repayment track record, and so forth. It may be noted that each branch office 34 includes a branch manager 38 and a workforce of agents 40 who are local to a region and are familiar with the people in the local community. The agents 40 interact with the consumers, provide adequate education about the processes, form groups, handle loan requests, disburse the loans, and collect the installments. In short, the agents 40 are the interface between the central microfinance institute 32 and the branch offices 34 and the consumers. The branch manager 38 oversees the operation of the branch, approves groups based on their understanding of the process, sanctions loans, keeps track of collections, handles money, and so forth.
[0036] Each agent 40 may be provided with a device 42 so as to increase their
operational efficiency. In one embodiment, the device 42 may include the exemplary adaptive loan appraisal system 10 (see FIG. 2) configured to aid the loan appraisal personnel in making a loan disbursement decision. The devices 42 assist the agents 40 in their fieldwork by automating certain processes of microfmancing. More particularly, in accordance with exemplary aspects of the present technique, the device 42 may be configured to aid loan appraisal personnel, such as the branch manager 38 in making a loan disbursement decision. It may be noted that the terms loan appraisal personnel and branch manager may be used interchangeably. In certain embodiments, the device 42 may include a portable device. However, in certain other embodiments, the device 42 may include a non-portable device, such as a computer (not shown in FIG. 3). Further, the device 42 may be configured to communicate with a central information system of

the microfmance entity (for example, MIS system 36 of the branch office 34) over a communication channel 44.
[0037] Furthermore, in accordance with exemplary aspects of the present
technique, the device 42 may include the exemplary adaptive loan appraisal system 10 for assisting the loan appraisal personnel 38 in making a loan disbursement decision. Additionally, in accordance with aspects of the present technique, it may be noted that the central microfmance institute 32, branch offices 34, and/or the portable device 42 may each include the adaptive loan appraisal system 10 for increasing the efficiency of the loan appraisal process.
[0038] FIG. 4 illustrates one embodiment 50 of the exemplary device 42 of
FIG. 3. As depicted in FIG. 4, the device 50 is shown as including the exemplary
adaptive loan appraisal system 10 (see FIG. 2) configured to aid the loan appraisal
personnel 38 (see FIG. 3) in making a loan disbursal decision. Additionally, the device
50 may also include a communication module 52. The communication module 52
communicates consumer specific data with a central information system of the
microfmance entity, such as the MIS system 36 (see FIG. 3) of the branch office 34 (see
FIG. 3), over the communication channel 44 (see FIG. 3). The consumer specific data
may include consumer details (for example consumer name, identification, address,
group association, and so forth) and associated transaction details (for example loan
approval and disbursement details such as sanctioned loan amount and date of
disbursement, installment and loan repayment details such as installment amount, period
of repayment, date of collections, payment details such as payment history, amount
collected till date, remaining amount, and so forth). It should be noted that the
communication channel 44 may be a secured communication channel and may include a
wired or a wireless communication channel such as Internet, local area network (LAN),
wide area network (WAN), data cables (for example RS232 cable, USB cable, and so
forth), GSM network, GPRS network, satellite network, and so forth.
[0039] Moreover, the device 50 may also include a transaction module 54. The
transaction module 54 may be configured to identify a consumer, such as a loan
applicant, access associated transaction details, process the associated transactions,
update and record the transaction details, and provide transaction details to the loan
applicant, the agent 40 (see FIG. 3), or the loan appraisal personnel 38.
[0040] Furthermore, the device 50 may be configured to identify a consumer,
such as the loan applicant. The identification of the loan applicant may be performed

based on consumer identification details provided via one or more peripheral devices 56. Input peripheral devices may include a finger print reader, an identification card reader, a keyboard, and/or a camera (not shown in FIG. 4). The loan applicant may provide his finger print impression on the finger print reader. Alternatively, the loan applicant may produce his identification card for being swiped in the card reader or for the identification number to be punched in through the keyboard. In certain embodiments, a camera may take picture of the loan applicant and/or the group for identification purposes. The associated transaction details of the identified loan applicant such as the amount that needs to be collected from loan applicant or the amount the needs to disbursed to the loan applicant may then be displayed to the agent 40 through a display device. The transaction module 54 may then process, update and record any performed transactions, generate a receipt of the performed transaction, and provide the consumer with a receipt via the one or more peripheral devices 56. Output peripheral devices may include a printer such as a thermal printer, a display, and/or a card write device (not shown in FIG. 4).
[0041] In certain embodiments, the transaction module 54 may be configured to
determine or compute transactions in-situ. For example, if a loan applicant requests a small loan, the transaction module 54 may communicate with the adaptive loan appraisal system 10 for initiating a loan disbursement decision. The adaptive loan appraisal system 10 may then provide a recommendation to assist the loan appraisal personnel 38 in approving or rejecting the current loan request. Based on the approval, the agent 40 may instantaneously provide the loan applicant a loan and update the records via use of the portable device 42. The recorded and updated data (consumer details and associated transaction details) may subsequently be transferred to the central information system or the MIS system 36 via the communication module 52. It should be noted that the transfer of data may be done at the end of the day when the agent 40 has finished his/her fieldwork over a wired or wireless network. Alternatively, the transfer of data may be done in real time or in batches at periodic intervals while the agent 40 is still in the field over a wired or a wireless network or via a phone-link (for example mobile operator network, telephone lines).
[0042] In certain embodiments, the branch office 34 (see FIG. 3) may track the
movement of the agents 40 for precautionary or dispute resolution purpose. The agents 40 sometimes carry substantial amounts of cash with them in remote locations and therefore need to be tracked. Moreover, the microfinance entities penalize the customers

even if they delay the installment payment by one day. The customers sometimes
complain that the agent 40 was not available at the identified collection point to collect
the installment on the appropriate day or at the appropriate time. Keeping track of the
agents' movement may therefore be important for the microfinance entities. The
portable device 42 may additionally include a position-sensing device 58 for receiving
positional information of the portable device 42. It should be noted that the position-
sensing device 58 may be adapted to track the location of the device 42 periodically or
continuously. The received or determined positional information may then be
communicated to the central information system 36 via the communication module 52
along with a date and time stamp. Moreover, the device 42 also captures the amount of
cash collected from customers or disbursed to customers at any given date and time.
Thus, the information regarding total cash available with the agent 42 may also be
relayed to the central information system 36 along with the positional information and
the date and time stamp. This enables the microfinance entity 30 to keep track of
whether the agent 40 reached the designated meeting location or the customers on time
and how much cash the agent 40 is carrying. In certain embodiments, the position-
sensing device 58 may include one or more of a GPS-based position sensing device, a
TV-based position-sensing device, a wireless access point network-based position-
sensing device, or a GSM network-based position-sensing device. These position-
sensing devices 58 utilize TV signals, GPS signals, GSM signals, or wireless network
signals for receiving or determining positional information of the portable device 42.
[0043] Additionally, the portable device 42 may generally include a processor
(not shown in FIG. 4), a memory (not shown in FIG. 4), and peripheral devices 56 (also referred as input/output devices) connected via a data pathway (for example buses, cables and so forth) (not shown in FIG. 4). The processor accepts instructions and data from the memory and performs various data processing fimctions of the device 42. Furthermore, the processor may include an arithmetic logic unit (ALU) (not shown in FIG. 4) that performs arithmetic and logical operations, and a control unit (not shown in FIG. 4) that extracts instructions from the memory and decodes and executes them, calling on the ALU when necessary.
[0044] The memory stores the various routines as well as a variety of data and
information computed by the various data processing functions of the device 42 or received from the input/output devices 56. These routines and data processing functions perform the above-discussed fiinctions of the communication module 52 and the

transaction module 54. The data may include, for example, transaction details consumer details related to the consumer. The memory generally includes a random-access memory (RAM) and a read-only memory (ROM); however, there may be other types of memory such as programmable read-only memory (PROM), erasable programmable read-only memory (EPROM) and electrically erasable programmable read-only memory (EEPROM). Also, the memory preferably contains an operating system that executes on the processor. The operating system performs basic tasks that include recognizing input, sending output to output devices, keeping track of files and directories and controlling various peripheral devices. The information in the memory might be conveyed to a human user through the peripheral devices 56, the data pathway, or in some other suitable maimer.
[0045] The peripheral devices 56 may act as an interface between a user, such
as the agent 40 or the loan applicant, and the device 42 and may include a keypad (not shown in FIG. 4) that enables the user to enter data and instructions into the device 42. Additionally, the peripheral devices 56 may include a display device (not shown in FIG. 4) that enables the user to view the available information and a printer (not shown in FIG. 4) that enables the user to print any data for reference. Further, the peripheral devices 56 enable the portable device 42 to acquire identification data. In particular, the peripheral device 56 may include a reader (not shown in FIG. 4) configured to acquire information from a readable device such as a transaction card and/or a mobile telephone. For example, the peripheral device 56 may include a card read/write device (not shown in FIG. 4) configured to read information from a transaction card and to write information on the transaction card when the transaction card is swiped or presented to the device 42. Alternatively, the peripheral device 56 may include a data cable, a Bluetooth port and/or an infrared port to read consumer identification details from the mobile telephone. The peripheral devices 56 may further include a finger print reader (not shown in FIG. 4) for scanning the finger print impression of the consumer and a camera (not shown in FIG. 4) to acquire the picture of the consumer. Alternatively, the peripheral devices 56 may include a retina scanner (not shown) or other similar consumer identification hardware. The finger print impression and the picture or other similar consumer identification information may be used for identification purposes or for preparing the loan application.
[0046] The device 42 may further include a communication port (not shown in
FIG. 4) such as a telephone, cable or wireless modem; a network card such as an

Ethernet adapter, local area network (LAN) adapter, integrated services digital network
(ISDN) adapter, or Digital Subscriber Line (DSL) adapter; a USB port; IEEE 1394 port;
and so forth, that enables the device 42 to access other devices, computers, and resources
on a communication channel 44. Li particular, in certain embodiments, the
communication port enables the portable device 42 to access the central information
system of a microfinance entity (for example the MIS system 36 of the branch office 34)
over the communication channel 44. As discussed above, the communication channel 44
may be a wired or a wireless communication networks or data cables.
[0047] An onboard rechargeable power source (for example rechargeable
battery) (not shown in FIG. 4) may be provided within the portable device 42 so that the
device 42 can function without needing constant supply of power in remote locations.
The rechargeable power source may be periodically charged through an external power
supply (not shown in FIG. 4) as and when the external power supply is available.
[0048] Moreover, the device 42 may also include a mass storage device (not
shown in FIG. 4) to allow the device 42 to retain large amounts of data permanently.
The mass storage device may include all types of disk drives such as floppy disks, hard
disks and optical disks, as well as tape drives that can read and write data onto a tape that
could include digital audio tapes (DAT), digital linear tapes (DLT), or other magnetically
coded media. It may be noted that the device 42 may also take the form of a hand-held
digital computer, personal digital assistant computer, notebook computer, and so forth.
[0049] As will be appreciated by one skilled in the art, a variety of techniques
may be employed to automate the process of microfinancing. In particular, as will be appreciated by those of ordinary skill in the art, control logic and/or automated routines for performing the techniques and steps described herein may be implemented by the portable device 42, either by hardware, software, or combinations of hardware and software. For example, suitable code may be accessed and executed by the processor to perform some or all of the techniques described herein. Similarly, application specific integrated circuits (ASICs) configured to perform some or all of the techniques described herein may be included in the processor.
[0050] For example, referring now to FIG. 5, exemplary control logic 60 for
microfinancing is depicted via a flowchart in accordance with aspects of the present technique. More particularly, a flowchart depicting a method for assisting loan appraisal personnel in making a loan disbursal decision is presented in FIG. 5. In certain embodiments, the method may be performed via a processor-based device, such as the

portable device 42 (see FIG. 3) including the exemplary adaptive loan appraisal system 10 (see FIG. 2).
[0051] The method starts at step 62 where input data may be accepted from a
loan applicant. As previously noted, the input module 12 (see FIG. 2) of the exemplary
adaptive loan appraisal system 10 may be used to accept input data from the loan
applicant. Further, the input data may include income details, expenditure details,
occupation of the loan appUcant, current loan request requirement details, purpose for the
current loan request, previous loan details, if any, or combinations thereof
[0052] Subsequently, at step 64, a range of gross surplus income of the loan
applicant may be computed. The step of computing the gross surplus income range may be better understood with reference to FIG. 6. Referring now to FIG. 6, a flowchart 70 depicting step 62 of FIG. 5 is illustrated. The gross siuplus income range may be computed based on an income potential, an expenditure pattern and occupation of the loan applicant. In certain embodiments, the method for computing the gross surplus income range may be performed via a processor-based device, such as the portable device 42 (see FIGs. 3-4). Specifically, the method for computing the gross surplus income range may be performed via use of the gross surplus module 14 (see FIG. 2) in the adaptive loan appraisal system 10.
[0053] In accordance with aspects of the present technique, the income potential
of the loan applicant may be determined based on income details 72 provided by the loan applicant and occupation 74 of the loan applicant as indicated by step 78. The input module 12 may be used to determine the income potential of the loan applicant, for example. The income details 72 of the loan applicant may include one to three sources of income that may be input as discrete values. As previously noted, the input module 12 may be used to accept input data from the loan applicant. Further, the input module 12 may be configured to determine an income potential of the loan applicant based on information related to the sources of income provided by the loan applicant. By way of example, the loan applicant may input a primary income PI, a secondary income SI and a tertiary income 77. It may be noted that sources of income may include a fixed income source, a variable income source, or both. As previously noted, the variable source of income may include a variable source of income with seasonality or a variable source of income without seasonality or both. The variable income with seasonality may be based on the season associated with the occupation of the loan applicant. If any source of income includes a variable source of income with seasonality, then income during a lean

season and a peak season may also be accepted as input. In addition, months
corresponding to the lean season and the peak season may also be accepted as input data.
[0054] Additionally, at step 80, an expenditure pattern of the loan applicant may
be determined based on expenditure details 76 provided by the loan applicant and the
occupation 74 of the loan applicant. By way of example, for every income source
provided by the loan applicant, an expenditure source may be provided as discrete
values. The input module 12 may be used to determine the expenditure pattern of the
loan applicant, for example. Here again, the expenditure details 76 of the loan applicant
may include one to three sources of expenditure that may be input as discrete values. As
previously noted, the input module 12 may be used to accept expenditure details from the
loan applicant. The input module 12 may be configured to determine an expenditure
pattern of the loan applicant based on information related to the sources of expenditure
provided by the loan applicant and the occupation 74 of the loan applicant. By way of
example, the loan applicant may input a primary expenditure PE, a secondary
expenditure SE and a tertiary expenditure TE. It may be noted that sources of
expenditure may include a fixed expenditure source, a variable expenditure source, or
both. Furthermore, it may be noted that the variable source of expenditure may include a
variable source of expenditure with seasonality, a variable source of expenditure without
seasonality, or both. The variable expenditure with seasonality may be based on the
season associated with the occupation of the loan applicant. If any source of expenditure
includes a variable source of expenditure with seasonality, then expenditure during a lean
season and a peak season may be accepted as input. In addition, months corresponding
to the lean season and the peak season may also be accepted as input data.
[0055] Subsequently, at step 82, a range of income of the loan applicant may be
determined. More particularly, a range of income corresponding to each of the sources of income PI, SI, and 77 may be determined. In one embodiment, the income module 12 may be used to determine the ranges corresponding to the various sources of income. By way of example, in one embodiment, for the primary source of income PI, a range of income may be determined as:
Range of PI=[RPI, , RPIj ] (1)
where

^^■

RPI,=

PI-

PI*

100

JJ

(2)

and

i?P/,=P/(l-0.15) RPl2=Pl{l + 0.05)

(3)
(4)

[0056] Similarly, for the secondary source of income 57, in one embodiment, a
range of income may be determined as:
Range of SI=[RSI, , RSI^ ] (5)
where

i?5/, =5/(1-0.15)

(6)

and RSI,=Sl{\+0.05) (7)
[0057] Moreover, for the tertiary source of income 77, in one embodiment, a
range of income may be determined as:

Range of TI=[RTI„RTI^]

(8)

where

7?r/, =7/(1-0.15)

(9)

and RTl2=Tl{\+0.05) (10)
[0058] Equations (1), (5) and (8) are representative of ranges of income
corresponding to fixed sources of income. It may be noted that if the sources of income include a variable source of income, then a range of income corresponding to the lean period and the peak period may be similarly determined.
[0059] Once a range of income for each of the sources of income is computed, a
range of expenditure corresponding to each of sources of expenditure may be computed, as indicated by step 84. By way of example, at step 84, for the primary source of expenditure PE, in one embodiment, a range of expenditure may be determined as:
RangeofPE=[RPE„RPE^] (11)

where
RPE, =PE{\-0.05) (12)
and RPE^=PE{l + 0A5) (13)
[0060] Moreover, for the secondary source of expenditure SE, in one
embodiment, a range of expenditure may be determined as:
Range of SE=[RSE^,RSE^] (14)
where
(15)
and (16)
[0061] Also, in one embodiment, a range of expenditure for the tertiary source
of expenditure TE, may be determined as:
RangeofTE=[RTE^,RTE^] (17)
where
^r^,=r£'(l-0.05) (18)
and RTE2=TE{\+0.\5) (19)
[0062] Once the ranges of income and expenditure corresponding to the sources
of income and expenditure are computed, the gross surplus income range may be determined based on the ranges of income and expenditure, as depicted by step 86. By way of example, in one embodiment, a gross primary income surplus range GPIS may be determined utilizing the range of primary income RPI and the range of primary expenditure RPE as:
(20)
[0063] Similarly, a gross secondary income surplus range GSIS may be
determined utilizing the range of secondary income RSI and the range of secondary expenditure RSE as:
(21)

[0064] Also, in one embodiment, a gross tertiary income surplus range GTIS
may be determined utilizing the range of tertiary income RTI and the range of tertiary expenditure RTE as:
GTIS = [{RTI, -RTE, ),{RTI, -RTE,)] (22)
[0065] Once a gross surplus income range corresponding to each of the sources
of income is determined (see equations (20), (21), (22)), an overall gross surplus income range GSI88 for the loan applicant may be computed as:
GSI=[GPIS+GSIS+GTIS] (23)
[0066] With returning reference to FIG. 3, subsequent to the computation of the
gross surplus income range 88 (see FIG. 6), a net surplus income range for the loan
applicant may be computed, as indicated by step 66. The computation of the net surplus
income range may be better understood with reference to FIG. 7.
[0067] Referring now to FIG. 7, a flowchart 90 depicting a method for
computing a range of net surplus income for the loan applicant is presented. More particularly, step 66 of FIG. 5 is illustrated in greater detail in FIG. 7. In certain embodiments, the method for computing the net surplus income range may be performed via a processor-based device, such as the portable device 42 (see FIGs. 3-4). Specifically, the method for computing the net surplus income range may be performed via use of the net surplus module 14 (see FIG. 2) in the adaptive loan appraisal system 10 (see FIG. 2).
[0068] In accordance with exemplary aspects of the present technique, the net
surplus income range may be computed based on the gross surplus income range 88 (see
FIG. 6), any previous loans 92 that the loan applicant has availed, and periodic
repayment installments 94 corresponding to the previous loans. More particularly,
amount borrowed corresponding to the previous loans 92 may be input to the net surplus
module 14 (see FIG. 2). Additionally, amounts corresponding to the periodic repayment
installments 94 and the terms of the previous loans may be input to the net surplus
module 14. Moreover, the source of the previous loans may also be input to the net
surplus module 14. By way of example, the sources of the previous loans may include a
bank, a microfinance institute, private lenders, or family and/or friends.
[0069] Based on the inputs provided to the net surplus module 14, a total
outstanding periodic repayment installment (EMI) for the loan applicant may be

computed, as indicated by step 96. By way of example, in one embodiment, the total outstanding periodic repayment installment M may be computed as:
(24)
where P is the loan amount or principal amount, i is the interest rate per annum, q is the number of payment periods per year, and n is the number of loan years.
[0070] In accordance with exemplary aspects of the present technique, the net
surplus income range may be computed based on the gross surplus income range 88 (see FIG. 6), and the total outstanding periodic repayment installments M (see Equation (24)), as depicted by step 98. More specifically, at step 98, the net surplus income range NSI 100 may be computed by subtracting the total outstanding periodic repayment installments M from the gross surplus income range GSI 88 (see Equation (23), FIG. 6) as follows:
(25)
[0071] With returning reference to FIG. 5, subsequent to the computation of the
net surplus income range 100 (see FIG. 7) of the loan applicant, the current loan request may be reviewed based on the input data and associated parameters to provide a recommendation regarding the current loan request, as depicted in step 68. As previously noted, the recommendation may include an approval or a rejection of the current loan request. Also, the loan recommender module 18 (see FIG. 2) may be employed to aid in reviewing the current loan request to provide a recommendation regarding the current loan request. Additionally, as indicated by step 69, the adaptive loan appraisal system 10 may also be configured to adaptively adjust one or more associated parameters based on the final decision of the loan appraisal personnel and will be described in greater detail with reference to FIG. 8. Steps 68-69 may be better understood with reference to FIG. 8.
[0072] Turning now to FIG. 8, a flowchart 110 depicting a method for
reviewing the current loan request to provide a recommendation regarding the current loan request is presented. More particularly, steps 68-69 of FIG. 5 are illustrated in greater detail in FIG. 8. In certain embodiments, the method for reviewing the current

loan request to provide a recommendation regarding the current loan request may be performed via a processor-based device, such as the portable device 42 (see FIG. 3). Specifically, the method for reviewing the current loan request to provide a recommendation for the current loan request may be performed via use of the loan recommender module 18 (see FIG. 2) in the adaptive loan appraisal system 10 (see FIG.
2).
[0073] In accordance with exemplary aspects of the present technique, in one
embodiment, the method for reviewing the current loan request to provide a
recommendation regarding the current loan request includes a three-stage verification
process. A first stage in the three-stage verification process includes computation of a
confidence value. Additionally, a second stage in the three-stage verification process
includes determination of a repayment capability of the loan applicant. Furthermore, a
third stage in the three-stage verification process includes computation of an experience
value. In accordance with aspects of the present technique, if the current loan request
satisfies requirements of all the three stages, then the current loan request may be
approved with a confidence level computed in the first stage. However, if the current
loan request fails in any of these stages, the current loan request may be rejected.
[0074] Furthermore, in accordance with other aspects of the present technique,
the adaptive loan appraisal system 10 (see FIG. 2) and specifically, the loan recommender module 18 (see FIG. 2) may be configured to review a current loan request 112 based on the input data 111 provided by the loan applicant and the net surplus income range 100 determined by the net surplus module 16 (see FIG. 2) to provide a recommendation regarding the current loan request 112 to the loan appraisal personnel, such as the branch manager 38 (see FIG. 3). The method starts at step 114, where a periodic repayment installment corresponding to the current loan request 112 may be computed. In one embodiment, the loan recommender module 18 may be configured to compute the periodic repayment installment corresponding to the current loan request 112. Furthermore, based on an income potential (for example the income potential computed at step 78 of FIG. 6) of the loan applicant and the variability of the income potential during a peak period and a lean period, the loan applicant may either be allocated an equal periodic repayment schedule or a non-equal periodic repayment schedule. More particularly, if the income of the loan applicant during the peak and lean periods falls within the net surplus income range 100, then the loan applicant is considered for equal periodic repayment installment based repayment, else the loan

applicant is assigned the non-equal periodic repayment installment option to repay the current loan request.
[0075] Subsequently, at step 116, a confidence level for the current loan request
may be computed. As previously noted, the term "confidence level" is used to refer to a
relative distance of an expected loan amount corresponding to the current loan request
from a maximum allowable value. Also, as previously noted with reference to FIG. 2,
the input module 12 may be configured to accept input data corresponding to the current
loan request from the loan applicant, where the input data may include income details,
expenditure details, occupation of the loan applicant, loan requirement details, purpose
for the current loan request, previous loan details, or combinations thereof
[0076] Moreover, the purpose for the current loan request may be divided into a
number of sub-items. In accordance with aspects of the present technique, minimum and maximum currency values corresponding to the sub-items may be retrieved from the Purpose Break Up Table, for example. In one embodiment, the loan recommender module 18 may be configured to automatically retrieve the minimum and maximum currency values corresponding to sub-items from the Purpose Break Up Table. More particularly, if an entry for the purpose for the current loan request exists in the Purpose Break Up Table, then all the sub-items associated with the purpose for the current loan request and the corresponding maximum and minimum currency requirements to accomplish each associated sub-item may be retrieved from the Purpose Break Up Table. A total maximum currency value Tmin may be computed based on the maximum currency values corresponding to all the sub-items. In a similar fashion, a total minimum currency value may be computed based on the minimum currency values
corresponding to all the sub-items. Subsequently, a check may be carried out to verify if the loan amount corresponding to the current loan request (expected loan amount) lies within the limits of the total minimum currency value Tmin and the total maximum currency value Tmax- If it is verified that the loan amount corresponding to the current loan request (expected loan amount) lies within the limits of the total minimum currency value Tmin and the total maximum currency value Tmax, then, by way of example, in one embodiment, a confidence level CL may be computed as:
(26)

[0077] At step 116, if the expected loan amount lies within the limits of the total
minimum currency value Tmin and the total maximum currency value Tmin, then the
confidence level CL may be set to TRUE. In one embodiment, if the expected loan
amount is less than the acceptance threshold value {Acceptance Threshold) of 0.6, then
the output of the first stage may be set to TRUE, thereby indicating that the expected
loan amount lies within the limits of the total minimum currency value Tmi„ and the total
maximum currency value Tmax- Further, if the expected loan amount is greater than the
total maximum currency value Tmax, then the confidence level CL may be set to FALSE.
Additionally, if the expected loan amount is less than the total minimum currency value
Tmin, then the confidence level CL may be set to FALSE. In accordance with aspects of
the present technique, if the expected loan amount is less than the total minimum
currency value Tmin or greater than the total maximum currency value Tmax, it may be
desirable to review the minimum and maximum currency values assigned to the sub-
items in the Purpose Break Up Table. Accordingly, the minimum and maximum
currency values assigned to the sub-items may be updated in the Purpose Break Up
Table. It may be noted that the loan recommender module 18 (see FIG. 1) may be
configured to adaptively adjust the minimum and maximum currency values assigned to
the sub-items in the Purpose Break Up Table. In one embodiment, the field information
of the loan appraisal personnel may be used to update the Purpose Break Up Table,
thereby facilitating "more informed" determination of the confidence levels, and
consequently the recommendation regarding the current loan request.
[0078] It may be noted that in certain embodiments, at the first stage, if it is
determined that the confidence level CL has a value set to TRUE, then control may be
passed on to the second stage. However, at the first stage, if it is determined that the
confidence level CL has a value set to FALSE, then the current loan request may be
rejected, and the rejection recommendation may be communicated to the loan appraisal
personnel. However, in certain other embodiments, control may be passed on to the
second stage independent of the value of the confidence level CL.
[0079] In accordance with aspects of the present technique, once the confidence
level CL is determined, repayment capability of the loan applicant may be determined as indicated by step 118. More particularly, a check may be carried out to verify if either the equal periodic repayment installment or the non-equal periodic repayment installment computed at step 114 is less than the net surplus income range 100 of the loan applicant (see FIG. 7). At step 118, if it is verified that the either equal periodic

repayment installment or the non-equal periodic repayment installment is less than the net surplus income range 100 of the loan applicant, then an output of the second stage is set to TRUE. However, at step 118, if it is verified that either the equal periodic repayment installment or the non-equal periodic repayment installment is greater than the net surplus income range 100 of the loan applicant, then an output of the second stage is set to FALSE. It may be noted that a TRUE output of the second stage is representative of an ability of the loan applicant to repay the expected loan amovint, while a FALSE output of the second stage is representative of an inability of the loan applicant to repay the expected loan amount.
[0080] Subsequently, at the third stage, an experience value may be determined,
as indicated by step 120. As used herein, the term "experience value" is representative of experience of the loan applicant in a general area associated with the purpose of the current loan request provided by the loan applicant as input data 111. Accordingly, at step 120, a check may be carried out to determine an experience of the loan applicant. At step 120, if it is determined that the experience of the loan applicant in the general area associated with the purpose of the current loan request is less than the experience threshold value Experience Threshold, then an output of the third stage may be set to FALSE and a recommendation for the current loan request may include a rejection of the current loan request. However, at step 120, if it is determined that the experience of the loan applicant in the general area associated with the purpose of the current loan request is greater than the threshold value Experience Threshold, then the output of the third stage may be set to TRUE.
[0081] Once the confidence level CL, the repayjment capability and the
experience value of the loan applicant are determined, the current loan request may be reviewed based on the determined confidence level CL, the repayment capability and the experience value of the loan applicant, as indicated by step 122. hi one embodiment, the loan recommender module 18 may be employed to review the current loan request based on the determine confidence level CL, the repayment capability and the experience value of the loan applicant. Specifically, in one embodiment, if at steps 116, 118 and 120, the outputs of all the three stages are determined to be TRUE, then a recommendation for the current loan request may be an approval of the current loan request. However, if at steps 116, 118 and 120, at least one output of the three stages is determined to be FALSE, then a recommendation for the current loan request may be a rejection of the current loan request.

[0082] Furthermore, at step 124, the determined recommendation regarding the
current loan request may be communicated to the loan appraisal persoimel. The loan appraisal personnel have the option of accepting or rejecting the recommendation of the adaptive loan appraisal system 10. The loan appraisal persoimel may arrive at a final decision regarding approving or rejecting the current loan request based on the recommendation of the adaptive loan appraisal system 10 and/or awareness of field information of the loan appraisal personnel. Furthermore, as indicated by step 126, the adaptive loan appraisal system 10 may also be configured to accept the final decision of the loan appraisal personnel. By way of example, if the final decision of the loan appraisal personnel includes a rejection of the recommendation of the current loan request, the adaptive loan appraisal system 10 may be configured to accept the final decision of the loan appraisal personnel, as indicated by step 126. Alternatively, if the final decision of the loan appraisal personnel includes an approval of the recommendation of the current loan request, the adaptive loan appraisal system 10 may be configured to accept the final decision of the loan appraisal personnel, as indicated by step 126. The final decision of the loan appraisal personnel may be stored in a data repository, such as the data repository 20 (see FIG. 2).
[0083] In accordance with further aspects of the present technique, the adaptive
loan appraisal system 10 may also be configured to adjust the one or more associated parameters with each loan appraisal, as indicated by step 128. More particularly, the adaptive loan appraisal system 10 may be configured to adjust the acceptance threshold value Acceptance Threshold, the acceptance limit value Acceptance Limit, the threshold increase level value Threshold Increase Level, and the experience threshold value Experience Threshold based on the final decision of the loan appraisal personnel for each loan appraisal. By implementing the adaptive loan appraisal system 10 to adaptively adjust the associated parameters based on the final decision of the loan appraisal corresponding to each loan appraisal, the adaptive loan appraisal system 10 may be configured to "adapt" itself and "learn" frorn each loan appraisal, thereby allowing the adaptive loan appraisal system 10 to provide more "informed" loan appraisal recommendations.
[0084] The system, device and techniques described in various embodiments
discussed above provide a system and method configured to review a current loan request and provide a recommendation regarding the current loan request based on fuzzy estimates provided by the customers and the loan appraisal personnel's understanding of

the market and the customers. With every acceptance and rejection, the adaptive loan appraisal system adaptively adjusts the associated parameters to make more "intelligent" decisions regarding future loan requests. Additionally, the device 42 is an application with a small footprint and is available in both PC and hand held (portable) format. Furthermore, this adaptive system adapts itself based on every loan appraisal being processed and hence advantageously off loads a lot of activities and time out of the loan appraisal personnel's schedule, thus freeing them up for other critical activities. The portable device may be carried into the fields, and hence faster appraisals may be performed.
[0085] While only certain features of the invention have been illustrated and
described herein, many modifications and changes will occur to those skilled in the art. It is, therefore, to be understood that the appended claims are intended to cover all such modifications and changes as fall within the true spirit of the invention.

CLAIMS:
1. An adaptive loan appraisal system for assisting loan appraisal personnel
in making a loan disbursement decision, the adaptive loan appraisal system comprising:
a loan recommender module configured to:
review a current loan request based on input data and one or more associated parameters to provide a recommendation for a current loan request; wherein the input data comprises income details, expenditure details, occupation of the loan applicant, loan requirement details, purpose for the current loan request, previous loan details, or combinations thereof; and
adaptively adjust the one or more associated parameters based on a final decision of the loan appraisal personnel regarding the current loan request.
2. The adaptive loan appraisal system of claim 1, further comprising:
an input module configured to accept input data from a loan applicant.
a gross surplus module configured to compute a range of gross surplus income based on an income potential, an expenditure pattern, and occupation of the loan applicant;
a net surplus module configured to determine a range of net surplus income based on the gross surplus income, previous loan details, and corresponding periodic repayment installments.
3. The adaptive loan appraisal system of claim 2, wherein the loan
recommender module is further configured to:
determine a periodic repayment installment for the current loan request;
compute a confidence level for the current loan request based on a minimum value and a maximum value associated with the purpose of the current loan request;
determine repayment capability of the loan applicant based on the range of net income surplus and the periodic repayment installment for the current loan request; and
determine an experience value corresponding to experience of the loan applicant in an area of the purpose of the current loan request.
4. The adaptive loan appraisal system of claim 3, wherein the loan
recommender module is configured to:

review the current loan request to provide a recommendation regarding the current loan request based on the computed confidence level, the determined repayment capability of the loan applicant, the determined experience value, and the one or more associated parameters; and communicate the recommendation for the current loan request to the loan appraisal personnel.
5. The adaptive loan appraisal system of claim 4, wherein the loan
recommender module is configured to:
accept a final decision of the loan appraisal personnel, wherein the final decision of the loan appraisal personnel is based upon awareness of field information of the loan appraisal personnel, and wherein the field information comprises information about the loan appraisal personnel's understanding about the purpose for which the current loan is sought, the occupation of the loan applicant, cash inflow and outflow of the loan applicant, peak and lean periods associated with the occupation of the loan applicant, socio-demographic status of the loan applicant and the loan appraisal personnel's understanding of the accuracy of the estimates provided by the loan applicant, or combinations thereof;
adaptively adjust the one or more associated parameters based on the final decision of the loan appraisal personnel; and
adaptively adjust the minimum value and the maximum value associated with the purpose of the current loan request.
6. A device for assisting loan appraisal personnel in making a loan
disbursement decision, the device comprising:
an adaptive loan appraisal system, comprising:
an input module for accepting input from a loan applicant;
a gross surplus module for computing a range of gross surplus income based on an income potential, an expenditure pattern, and occupation of the loan applicant;
a net surplus module for determining a range of net surplus income based on the range of gross surplus income, previous loan details and corresponding periodic repayment installments;
a loan recommender module for:
reviewing a current loan request based on input data and one or more associated parameters to provide a recommendation for a current loan request; and

adaptively adjusting the one or more associated parameters based on a final decision of the loan appraisal personnel regarding the current loan request.
7. The device of claim 6, wherein the device is a portable device or a fixed device.
8. The device of claim 6, wherein the adaptive loan appraisal system is further configured to:
compute a confidence level for the current loan request based on a minimum value and a maximum value associated with a purpose of the current loan request;
determine repayment capability of the loan applicant based on the range of net income surplus and a periodic repayment installment for the current loan request;
determine an experience value corresponding to experience of the loan applicant in an area of the purpose of the current loan request;
review the current loan request to provide a recommendation for the current loan request based on the computed confidence level, the determined repayment capability of the loan applicant, the determined experience value and the associated parameters;
communicate the recommendation to the loan appraisal personnel;
accept a final decision of the loan appraisal personnel;
adaptively adjust one or more parameters based on the final decision of the loan appraisal personnel; and
adaptively adjust the minimum value and the maximum value associated with the purpose of the current loan request.
9. A method for assisting loan appraisal personnel in making a loan
disbursement decision, the method comprising:
accepting input data from a loan applicant;
reviewing a current loan request based on the input data and one or more associated parameters to provide a recommendation for a current loan request; and
adaptively adjusting the one or more associated parameters based on a final decision of the loan appraisal personnel regarding the current loan request.
10. The method of claim 9, wherein accepting input data from a loan
applicant comprises accepting income details, expenditure details, occupation, loan

requirement details, previous loan details, or combinations thereof fi-om the loan applicant.
11. The method of claim 10, further comprising:
computing a range of gross surplus income based on an income potential, an expenditure pattern, and an occupation of the loan applicant; and
determining a range of net surplus income based on the range of gross surplus income, previous loan details, and corresponding periodic repayment installments.
12. The method of claim 11, further comprising:
computing a confidence level for the current loan request based on a minimum value and a maximum value associated with the purpose of the current loan request;
determining repayment capability of the loan applicant based on the range of net income surplus and a periodic repayment installment for the current loan request; and
determining an experience value corresponding to experience of the loan applicant in an area of the purpose of the current loan request.
13. The method of claim 12, further comprising:
reviewing the current loan request to provide a recommendation regarding the current loan request based on the computed confidence level, the determined repayment capability of the loan applicant, the determined experience value, and the one or more associated parameters; and
communicating the recommendation for the current loan request to the loan appraisal persormel.
14. The method of claim 13, further comprising:
accepting a final decision of the loan appraisal personnel, wherein the final decision of the loan appraisal personnel is based upon awareness of field information of the loan appraisal personnel, and wherein the field information comprises information about the loan appraisal persormel's understanding about the purpose for which the current loan is sought, the occupation of the loan applicant, cash inflow and outflow of the loan applicant, peak and lean periods associated with the occupation of the loan applicant, socio-demographic status of the loan applicant and the loan appraisal

personnel's understanding of the accuracy of the estimates provided by the loan applicant, or combinations thereof;
adaptively adjusting the one or more associated parameters based on the final decision of the loan appraisal personnel; and
adaptively adjusting the minimum value and the maximum value associated with the purpose of the current loan request.
15. A method for assisting loan appraisal personnel in making a loan disbursement decision, the method comprising:
accepting input data from a loan applicant via an input module;
computing a range of gross surplus income based on an income potential, an expenditure pattern, and knowledge of occupation of the applicant via a gross surplus module;
determining a range of net surplus income based on a range of gross surplus income of the loan applicant, previous loans and corresponding periodic installments via a net surplus module;
reviewing a current loan request based on the input data and associated parameters to provide a recommendation for the current loan request via a loan recommender module;
adaptively adjusting a minimum value and a maximum value associated with a purpose of the current loan request via a loan recommender module; and
adaptively adjusting the one or more associated parameters based on a final decision of the loan appraisal personnel regarding the current loan request via the loan recommender module.