DESC:TECHNICAL FIELD
[0001] The present invention relates to a system and a method for live fraud detection during a computer based online examination in an unknown and non-secured environment. The invention, more particularly, relates to generally to an infrastructure for securely conducting computer-based examination/test (CBT) in a non-trusted network environment with live fraud detection with a customized system and method.
BACKGROUND
[0002] With advancements in computer technologies, several examinations are now being conducted digitally through computer-based tests. Currently, the problem occurs to securely conduct a computer-based examination/test (CBT) in a non-trusted network environment. The online conduction of examination on behalf of customers are for the purposes such as recruitment, appraisal, admission, certification and the like. The customers include organizations like educational boards, government and private entities. To conduct the exam, third-party venues such as colleges and schools are contracted for their computer lab infrastructure. This infrastructure is available for a very limited time for pre/during/post the examination and with unknown / inappropriate in-built security protocols, such infrastructure is rendered untrustworthy from both internal and external perspective. Further, the people hired for conducting this examination at the ground level have limited technology acumen, driving the need for a reliable system.
[0003] These alien environment coupled with the expectation to get real time results warrant the creation of a system that is quick, versatile and end-user friendly in deployment. The existing technologies are not able to provide a real time solution in an untrustworthy environment.
[0004] Various conventional technologies have drawbacks such as high costs, technical challenges, accessibility issues, dependence on physical characteristics. Also, the candidates may face technical issues such as software incompatibilities, inadequate hardware, or unreliable internet connections, which can hinder their ability to complete the examination successfully. Various biometric systems and keystroke analysis depend on consistent physical characteristics or behaviours, which may vary due to stress, illness, or technical issues, leading to authentication errors or false suspicions.
[0005] In view of the above deficiencies mentioned in the conventional approaches, there is a need to have a technical solution to ameliorate one or more problems of the conventional problems or to at least provide a solution to provide a method and a system which analyzes the behaviour of candidates taking such computer based online examinations in real time and provides a real time fraud analysis.

SUMMARY
[0006] This summary is provided to introduce concepts related to a system and method designed for the live detection of fraudulent activities during the administration of computer-based online examinations conducted within an unknown and non-secured network environment. More specifically, the invention relates to an innovative infrastructure that facilitates the secure execution of computer-based tests (CBT) in environments where network security cannot be guaranteed, integrating a bespoke system and method for real-time fraud detection. This infrastructure is adept at ensuring the integrity and credibility of online examinations by employing advanced monitoring and analysis techniques to identify and mitigate instances of malpractice instantaneously, thereby establishing a robust framework for conducting high-stakes assessments in virtually any network setting, irrespective of its inherent security posture. This summary is neither intended to identify essential features of the present invention nor is it intended for use in determining or limiting the scope of the present invention.
[0007] For example, various embodiments herein may include one or more systems for live fraud detection during a computer based online examination in an unknown and non-secured environment and methods thereof. In one of the embodiments, the system for live fraud detection during a computer based online examination in an unknown and non-secured environment is disclosed. This system includes a Master that serves as a database and a Node which is in communication with the Master. The Node is connected to a custom add-on hardware via a peripheral component interconnect express (PCIe) bus, wherein the Node is configured to process telemetry data received from a plurality of candidate terminals for detecting live fraud. The candidate terminals are computer machines used by a plurality of candidates present in a venue for the online examination. This Node further includes an answer pattern analyzer which is configured to analyse the answering pattern of the plurality of candidates to generate decision in real time. A screenshot processor which is configured to capture a screenshot of the candidate terminals for further processing. A candidate mood detector which is configured to analyse the candidate behaviour during the online examination. An image processor which is configured to analyse the data captured by a skeletal movement tracker and a camera image processor in real time. The skeletal movement tracker is configured to track the skeletal movement of the plurality of candidates in real time and the camera image processor is configured to perform image analysis of the plurality of candidates at their respective candidate terminals. Further, the Node is configured to detect live fraud based on the analysed answering pattern of the plurality of candidates, the captured screenshot of the candidate terminals, the analysed candidate behaviour, the tracked skeletal movement and the generated image analysis of the plurality of the candidates.
[0008] Further, in another embodiment of the present invention, the method is implemented by a live fraud detection system, wherein the system comprises a Master and a Node in communication with the Master. This Node is connected to a custom add-on hardware. The method includes processing a telemetry data coming from a plurality of candidate terminals for detecting live fraud by the Node. These candidate terminals are computer machines used by a plurality of candidates present in a venue for the online examination. Further, the method includes analysing the answering pattern of the plurality of candidates to generate decision in real time by an answer pattern analyzer. Further, capturing a screenshot of the candidate terminals for further processing by a screenshot processor and analysing the candidate behaviour during the online examination by a candidate mood detector. The method further includes analysing, by an image processor, the data captured by a skeletal movement tracker and a camera image processor in real time. This tracking of the skeletal movement of the plurality of candidates in real time is performed by the skeletal movement tracker and the image analysis of the plurality of candidates at their respective candidate terminals is performed by the camera image processor. Further, the method includes detecting live fraud based on the analysed answering pattern of the plurality of candidates, the captured screenshot of the candidate terminals, the analysed candidate behaviour, the tracked skeletal movement and the generated image analysis of the plurality of the candidates.
BRIEF DESCRIPTION OF ACCOMPANYING DRAWINGS
[0009] The detailed description is described with reference to the accompanying figures. In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. The same numbers are used throughout the drawings to reference like features and modules.
[0010] Figure 1 block diagram representing a Node connected to a customized add-on hardware i.e., MOS-D via PCIe bus for live fraud detection, according to an exemplary implementation of the present invention.
[0011] Figure 2 illustrates a block diagram representing the communication of the Node Core with the MOS-D using an RPC API over the PCIe bus, according to an exemplary implementation of the present invention.
[0012] Figure 3 is a flowchart of a method for live fraud detection during a computer based online examination in an unknown and non-secured environment, according to an exemplary implementation of the present invention.
[0013] It should be appreciated by those skilled in the art that any block diagrams herein represent conceptual views of illustrative apparatuses embodying the principles of the present invention. Similarly, it will be appreciated that any flow charts, flow diagrams, and the like represent various processes which may be substantially represented in computer readable medium and so executed by a computer or processor, whether or not such computer or processor is explicitly shown.
DETAILED DESCRIPTION
[0014] The various embodiments of the present invention provides a system and method designed for the live detection of fraudulent activities during the administration of computer-based online examinations conducted within an unknown and non-secured network environment. More specifically, the invention relates to an innovative infrastructure that facilitates the secure execution of computer-based tests (CBT) in environments where network security cannot be guaranteed, integrating a bespoke system and method for real-time fraud detection. This infrastructure is adept at ensuring the integrity and credibility of online examinations by employing advanced monitoring and analysis techniques to identify and mitigate instances of malpractice instantaneously, thereby establishing a robust framework for conducting high-stakes assessments in virtually any network setting, irrespective of its inherent security posture.
[0015] In the following description, for purpose of explanation, specific details are set forth in order to provide an understanding of the present invention. It will be apparent, however, to one skilled in the art that the present invention may be practiced without these details. One skilled in the art will recognize that embodiments of the present invention, some of which are described below, may be incorporated into a number of systems.
[0016] However, the methods and systems are not limited to the specific embodiments described herein. Further, structures and devices shown in the figures are illustrative of exemplary embodiments of the present invention and are meant to avoid obscuring of the present invention.
[0017] Furthermore, connections between components and/or modules within the figures are not intended to be limited to direct connections. Rather, these components and modules may be modified, re-formatted or otherwise changed by intermediary components and modules.
[0018] In accordance with one embodiment of the present invention, a system is disclosed for the real-time detection of fraudulent activities during the administration of computer-based online examinations within environments characterized by unknown or non-secured network conditions. The system comprises a Master, functioning as a centralized database, and a Node, which establishes communication with the Master. The Node is interconnected to specialized add-on hardware through a Peripheral Component Interconnect Express (PCIe) bus, and is specifically configured to process telemetry data emanating from a plurality of candidate terminals. These terminals represent the computer devices operated by candidates situated within a designated examination venue. Integral to the Node is an answer pattern analyzer, tasked with evaluating the response patterns of the candidate cohort to facilitate immediate decision-making. Additionally, the system incorporates a screenshot processor, designed for the acquisition and subsequent analysis of screenshots from the candidate terminals. A candidate mood detector is also included, with a mandate to assess behavioral patterns of candidates during the examination process. The system further integrates an image processor capable of interpreting data obtained from both a skeletal movement tracker and a camera image processor in real-time. The skeletal movement tracker is charged with monitoring the candidates' skeletal movements instantaneously, while the camera image processor undertakes the analysis of images captured of candidates at their respective terminals. Moreover, the Node is adept at identifying instances of live fraud through a comprehensive analysis encompassing the evaluated answering patterns, obtained screenshots, assessed candidate behaviors, monitored skeletal movements, and the conducted image analysis of the candidate assembly, thereby ensuring the integrity of the examination process in real-time.
[0019] In yet another embodiment of the present invention, a method is articulated for implementation by a system configured for the detection of fraudulent activities in real-time, during the administration of computer-based online examinations. The system is architecturally composed of a Master and a Node, wherein the Node maintains communicative linkage with the Master and is augmented through the connection to custom-designed add-on hardware. The method initiates with the Node's processing of telemetry data sourced from an array of candidate terminals, which are essentially computing devices deployed by numerous candidates congregated within a specified examination venue, for the purpose of identifying instances of live fraud. The method further includes analyzing the answering patterns wherein an answer pattern analyzer is engaged to scrutinize the response behaviors exhibited by the candidate cohort, with the objective of facilitating the generation of decisions in a real-time context. Further, the method includes capturing screenshots wherein a screenshot processor is utilized to execute the capture of screenshots from the candidate terminals, earmarking them for subsequent processing. The method further includes evaluating candidate behavior wherein the candidate mood detector undertakes the analysis of candidate dispositions during the examination, providing insights into behavioral patterns. The method further includes real-time data analysis wherein an image processor is tasked with the analysis of data harvested by both a skeletal movement tracker and a camera image processor in a concurrent fashion. The skeletal movement tracker is specifically designed to monitor the dynamic skeletal movements of the candidates in real-time, while the camera image processor is responsible for conducting a comprehensive image analysis of the candidates stationed at their respective terminals.
[0020] The culmination of the method involves the Node's detection of live fraud, predicated on a synthesis of analyzed data points including the candidates' answering patterns, screenshots from the candidate terminals, assessments of candidate behavior, tracking of skeletal movements, and the resultant image analyses of the candidate assembly. This multi-faceted approach ensures a robust framework for the identification and mitigation of fraudulent activities during the conduct of computer-based online examinations.
[0021] In another embodiment of the present invention, the telemetry data includes the keystrokes, candidate images, screenshots and connected devices, said telemetry data collected from the candidate terminals is processed on the custom add-on hardware and stored on the Master via the Node, wherein the custom add-on hardware is a MOS-D.
[0022] In accordance with the present invention, the Node is specifically configured to execute a series of operations aimed at enhancing the security and integrity of computer-based online examinations. The configuration of the Node includes enabling Preboot execution Environment (PXE) Protocol wherein the Node is equipped to activate and utilize the Preboot Execution Environment (PXE) protocol within the network infrastructure. This capability allows for the network-based booting of a computer system, enabling the Node to initiate and control the candidate terminals from a remote location, even before the operating system is loaded. This function is instrumental in establishing a standardized examination environment across all candidate terminals, ensuring that only authorized software and systems are operational during the examination. Further, the Node includes analyzing telemetry data. The Node is adept at processing and analyzing telemetry data received from the candidate terminals. This data, which encompasses a wide array of metrics related to the operational status, behavioral patterns, and interaction data of the examination software on the candidate terminals, is scrutinized for indicators of live fraud. Through sophisticated algorithms and pattern recognition techniques, the Node evaluates this telemetry data in real-time to identify anomalies or behaviors consistent with fraudulent activities, thereby maintaining the examination's integrity. The Node further includes providing Address Resolution Protocol (ARP) Enforcement to safeguard against Internet Protocol (IP) spoofing, a technique often employed to gain unauthorized access or disrupt network communications, the Node implements Address Resolution Protocol (ARP) enforcement. This security measure ensures that all IP addresses within the examination network are authenticated and verified, preventing malicious entities from impersonating other devices on the network. By enforcing ARP rules, the Node effectively mitigates the risk of IP spoofing attacks, thereby preserving the secure and fair conduct of online examinations.
[0023] Through these configurations, the Node plays a pivotal role in the secure and efficient management of computer-based online examinations, ensuring that the examinations are conducted in a manner that is both secure against fraudulent activities and resilient to common network security threats.
[0024] In accordance with an embodiment of the present invention, the screenshot processor is configured to capture the screenshot of the candidate terminals for surveillance. This functionality is designed to systematically record visual evidence of the candidates' activities and the content displayed on their screens during the examination period. Such captures facilitate a comprehensive review and monitoring process, enabling the detection of prohibited behaviours or unauthorized content, thus reinforcing the integrity and security of the examination environment.
[0025] In accordance with an embodiment of the present invention, the camera image processor is configured to perform image analysis of the plurality of candidates at their respective candidate terminals based on facial recognition. This configuration enables the precise identification and verification of candidates participating in the examination by analyzing facial features in real-time. The incorporation of facial recognition algorithms allows for the continuous monitoring of the examination environment, ensuring that only authorized candidates are present and actively engaged in the examination process. This technology plays a pivotal role in maintaining the integrity and security of the examination by preventing impersonation and other forms of fraudulent activities.
[0026] It should be noted that the description merely illustrates the principles of the present invention. It will thus be appreciated that those skilled in the art will be able to devise various arrangements that, although not explicitly described herein, embody the principles of the present invention. Furthermore, all examples narrated herein are principally intended expressly to be only for explanatory purposes to help the reader in understanding the principles of the invention and the concepts contributed by the inventor to furthering the art and are to be construed as being without limitation to such specifically recited examples and conditions. Moreover, all statements herein reciting principles, aspects, and embodiments of the invention, as well as specific examples thereof, are intended to encompass equivalents thereof.
[0027] The objective of the present invention is to securely conduct a computer-based examination/test (CBT) in a non-trusted network environment with live fraud detection with a customized system and method.
[0028] In an embodiment of the present invention, a system for live fraud detection during computer based online examinations and a method thereof is disclosed.
[0029] Figure 1 illustrates a block diagram representing a Node connected to a customized add-on hardware i.e., MOS-D via PCIe bus for live fraud detection, according to an exemplary implementation of the present disclosure. For live fraud detection, considerable amount of processing power is required. Hence, Node has a custom PCIe add-on hardware named MOS-D which processes a plurality of features. One of the features is processing telemetry coming from candidate terminals i.e., personal computers. Another feature is processing of real time images/photos of the candidates from the candidate terminals and processing screenshots of the candidate terminals taken in real time. It also provides intelligence on the candidate behaviour during the examination.
[0030] Figure 2 illustrates a block diagram representing the communication of the Node Core (202) with the MOS-D using a Remote Procedure Call (RPC) API (204) over the PCIe bus, according to an exemplary implementation of the present disclosure. The processing of the answer pattern and the mood detection of the candidate taking the examination forms an integral part of the insights shared regarding a particular candidate. The answer pattern analyzer (206) comprises of a HID Processor (210) and a screenshot processor (212). The image processor (214) forms the computer vision based on the analysis of the captured data, wherein the data includes the skeletal movement of the candidate captured by the skeletal movement tracker (216) and the camera image of the candidate during the examination captured by the camera image processor (218). The telemetry data collected from the candidate terminals is processed on the MOS-D and the insights are recorded on the Master. The telemetry data includes keystrokes, candidate photos, screenshots, connected devices and the like.
[0031] Figure 3 is a flowchart of a method for live fraud detection during a computer based online examination in an unknown and non-secured environment, according to an exemplary implementation of the present invention.
[0032] At step 302, processing the telemetry data coming from a plurality of candidate terminals for detecting live fraud by the Node (202). In another embodiment, the Node is configured to process telemetry data received from a plurality of candidate terminals for detecting live fraud, wherein the candidate terminals are computer machines used by a plurality of candidates present in a venue for the online examination.
[0033] At step 304, analysing the answering pattern of the plurality of candidates to generate decision in real time by an answer pattern analyzer (206). In another embodiment, an answer pattern analyzer (206) configured to analyse the answering pattern of the plurality of candidates to generate decision in real time.
[0034] At step 306, capturing a screenshot of the candidate terminals for further processing by a screenshot processor (212). In another embodiment, a screenshot processor (212) configured to capture a screenshot of the candidate terminals for further processing.
[0035] At step 308, analysing the candidate behaviour during the online examination by a candidate mood detector (208). In another embodiment, a candidate mood detector (208) configured to analyse the candidate behaviour during the online examination.
[0036] At step 310, analysing the data captured by a skeletal movement tracker (216) and a camera image processor (218) in real time, by an image processor (214). In another embodiment, an image processor (214) configured to analyse the data captured by a skeletal movement tracker (216) and a camera image processor (218) in real time. The skeletal movement tracker (216) is configured to track the skeletal movement of the plurality of candidates in real time and the camera image processor (218) is configured to perform image analysis of the plurality of candidates at their respective candidate terminals.
[0037] At step 312, detecting live fraud based on the analysed answering pattern of the plurality of candidates, the captured screenshot of the candidate terminals, the analysed candidate behaviour, the tracked skeletal movement and the generated image analysis of the plurality of the candidates. In another embodiment, the Node (202) is further configured to detect live fraud based on the analysed answering pattern of the plurality of candidates, the captured screenshot of the candidate terminals, the analysed candidate behaviour, the tracked skeletal movement and the generated image analysis of the plurality of the candidates.
[0038] In another embodiment of the present invention, live fraud detection in a computer-based examination (CBT) infrastructure is provided. Further, Over-The-Air (OTA) update of the artificial intelligence models are trained from aggregated data across venues.
[0039] In another embodiment of the present invention, the image processor can provide a real time tracking of candidate movement through skeletal movement and image analysis using facial recognition of the candidate at his/her terminal. The answer pattern analyzer uses advanced artificial intelligence capabilities to predict the answering pattern as well as screenshot of the candidate terminal to make a real time decision making.
[0040] In another embodiment of the present invention, the present disclosure provides a solution in real time in an untrustworthy environment wherein the candidate photo is processed from the candidate terminal in real time, and it is further analyzed. Also, the screenshots of the candidate terminals are processed, and the sanity of the candidate terminal is checked. Further, the system provides intelligence on the candidate behaviour during the examination. This includes analyzing the answering pattern of the candidate in real time and doing a real time fraud analysis.
[0041] It should be noted that the description merely illustrates the principles of the present invention. It will thus be appreciated that those skilled in the art will be able to devise various arrangements that, although not explicitly described herein, embody the principles of the present invention. Furthermore, all the used cases recited herein are principally intended expressly to be only for explanatory purposes to help the reader in understanding the principles of the invention and the concepts contributed by the inventor(s) to furthering the art and are to be construed as being without limitation to such specifically recited used cases and conditions. Moreover, all statements herein reciting principles, aspects, and embodiments of the invention, as well as specific examples thereof, are intended to encompass equivalents thereof.
,CLAIMS:
1. A system for live fraud detection during a computer based online examination in an unknown and non-secured environment, said system comprising:
a Master serving as a database; and
a Node (202) in communication with the Master, said Node (202) connected to a custom add-on hardware via a peripheral component interconnect express (PCIe) bus, said Node (202) is configured to process telemetry data received from a plurality of candidate terminals for detecting live fraud, wherein the candidate terminals are computer machines used by a plurality of candidates present in a venue for the online examination, said Node (202) further comprises:
an answer pattern analyzer (206) configured to analyse the answering pattern of the plurality of candidates to generate decision in real time;
a screenshot processor (212) configured to capture a screenshot of the candidate terminals for further processing;
a candidate mood detector (208) configured to analyse the candidate behaviour during the online examination;
an image processor (214) configured to analyse the data captured by a skeletal movement tracker (216) and a camera image processor (218) in real time, wherein the skeletal movement tracker (216) is configured to track the skeletal movement of the plurality of candidates in real time and the camera image processor (218) is configured to perform image analysis of the plurality of candidates at their respective candidate terminals;
wherein the Node (202) is further configured to detect live fraud based on the analysed answering pattern of the plurality of candidates, the captured screenshot of the candidate terminals, the analysed candidate behaviour, the tracked skeletal movement and the generated image analysis of the plurality of the candidates.

2. The system as claimed in claim 1, wherein the telemetry data includes the keystrokes, candidate images, screenshots and connected devices, said telemetry data collected from the candidate terminals is processed on the custom add-on hardware and stored on the Master via the Node, wherein the custom add-on hardware is a MOS-D.

3. The system as claimed in claim 1, wherein the Node is configured to:
enable a Preboot Execution Environment (PXE) protocol in the network;
analyse the telemetry data received from the candidate terminals for detecting live fraud and;
provide Address Resolution Protocol (ARP) enforcement for protecting from Internet Protocol (IP) spoofing.

4. The system as claimed in claim 1, wherein the screenshot processor (212) is configured to capture the screenshot of the candidate terminals for surveillance.

5. The system as claimed in claim 1, wherein the camera image processor (218) is configured to perform image analysis of the plurality of candidates at their respective candidate terminals based on facial recognition.

6. A method for live fraud detection during a computer based online examination in an unknown and non-secured environment, said method implemented by a live fraud detection system, said system comprising a Master and a Node (202) in communication with the Master, said Node (202) connected to a custom add-on hardware;
the method comprising:
processing, by the Node (202), telemetry data coming from a plurality of candidate terminals for detecting live fraud, wherein the candidate terminals are computer machines used by a plurality of candidates present in a venue for the online examination;
analysing, by an answer pattern analyzer (206), the answering pattern of the plurality of candidates to generate decision in real time;
capturing, by a screenshot processor (212), a screenshot of the candidate terminals for further processing;
analysing, by a candidate mood detector (208), the candidate behaviour during the online examination;
analysing, by an image processor (214), the data captured by a skeletal movement tracker (216) and a camera image processor (218) in real time, wherein the tracking of the skeletal movement of the plurality of candidates in real time is performed by the skeletal movement tracker (216) and the image analysis of the plurality of candidates at their respective candidate terminals is performed by the camera image processor (218);
detecting live fraud based on the analysed answering pattern of the plurality of candidates, the captured screenshot of the candidate terminals, the analysed candidate behaviour, the tracked skeletal movement and the generated image analysis of the plurality of the candidates.

7. The method as claimed in claim 6, wherein the method comprises:
enabling a Preboot Execution Environment (PXE) protocol in the network;
analysing the telemetry data received from the candidate terminals for detecting fraud and;
providing Address Resolution Protocol (ARP) enforcement for protecting from Internet Protocol (IP) spoofing.

8. The method as claimed in claim 6, wherein capturing the screenshot of the candidate terminals for surveillance.

9. The method as claimed in claim 6, wherein the image analysis of the plurality of candidates at their respective candidate terminals is performed based on facial recognition.