Claims:[0032] CLAIMS:
We claim
1. A system to measure the effectiveness and performance of commercial rules and learnt rules that recommend content to user (s), the system comprising:
a. an input module (102), wherein the input module (102) in a mobile device (101) allows the user (s) to login and type the required content therein;
b. a setup module (104) in a server (103) allows an admin to create a plurality of rules, wherein the rules qualify the typed content by the user (s);
c. a database module (105), wherein the database module (105) stores, processes and paginates the created rules to provide one or more recommendations based on the typed content thereby;
d. a data lake module (106) in the database module (105) analyzes the recommended contents based on an intelligent decision making using the crated rules, wherein each recommendation is assigned and assembled with a rule identification therein; and
e. a display module displays the effective and relevant recommendations with their respective rule identification to the user (s) through the mobile device therein.

2. The system as claimed in claim 1, wherein the system (100) tracks and marks the interactions of the user (s) against the recommendations with their respective rule identification that are sent to the user (s) and used by the user (s).

3. A method to measure the effectiveness and performance of commercial rules and learnt rules that recommend content to user (s), the method comprising the steps of:
a. logging into an input module in a mobile device and allowing the user (s) to type the required content therein (201);
b. creating a plurality of rules by an admin at a server (202);
c. storing the rules into a data base (203);
d. assembling the data in the server by using the data stored in the database and sending the data to user (s) at the input module, wherein every data point is assigned with a rule identification (ID’s) forming the recommendations (204);
e. storing the recommendations in to a data lake (205);
f. receiving the recommendations by the user (s) and interacting with the recommendations, wherein the interactions are sent to the server with their rule IDs (206);
g. marking the rule IDs against the recommendations in the data lake (207);
h. analyzing the data in the data lake and measuring the performance of the rules (208); and
i. displaying the effective and relevant recommendations with their respective rule identification to the user (s) through the mobile device therein (209).

4. The method as claimed in claim 1, wherein the method tracks and marks the interactions of the user (s) against the recommendations with their respective rule identification that are sent to the user (s) and used by the user (s). , Description:[0001] PREAMBLE TO THE DESCRIPTION:
[0002] The following specification particularly describes the invention and the manner in which it is to be performed:
[0003] DESCRIPTION OF THE INVENTION:
[0004] Technical field of the invention
[0005] The present invention relates to a system and method to measure effectiveness of the recommendations accurately. More particularly, the invention relates to a system to measure the effectiveness and performance of commercial rules and the learnt rules that recommend content to user (s).
[0006] Background of the invention
[0007] Recommendation platforms are part of data filtering engine that try to predict the 'rating' or 'preference' that a user would give to a product. Recommender systems have become quite common these days, and are used in a plurality of applications such as movies, music, news, books, research articles, search queries, social tags, and products in general. However, there are also recommender systems for restaurants, financial services, life insurance, persons (online dating), and Twitter followers etc.
[0008] Online shopping has increased radically in today’s world. The race among online merchants to provide the best user experience has become difficult. The online vendors attempt to implement various techniques to increase the sales. Today one of the techniques used is “recommendation”, in which the user when looking for particular product will be provided with a recommendation. Generally, the recommendation engine detects the products that were sold to other users along with the products that are currently being viewed by the user. However, in existing systems these recommendations are irrelevant, obsolete or replication due to inefficient measurement of the user’s interaction.
[0009] Further, the search engines play very important role in today’s online market/internet for interactive marketing. The search engine performs a search using a key word entered by the user and displays the best matching content to the user with respect to the entered key words and earlier activities. The effectiveness of the relevant data is beneficial in interactive marketing, in which the commercial search engines display the content that enables the sales increment of business entity. However, the existing search engines fails to measure the effectiveness of the user activities accurately, which in business perspective leads to loosing opportunities for advertising profits, in the case of e-Marketing sites, e-Commerce and e-Travel.
[0010] Generally, the recommendations in an application is a common method to push relevant content to the user in a content based application, but the effectiveness of the recommendations in the existing system cannot be measured accurately. This hinders its continuous improvement and reduces the user’s effective engagement with the system.
[0011] Currently, many learning methods employ the 80-20 rule of measurement. Data scientists and learning engines run through the data set and build a polynomial equation that specifies rules on the 80 percent of the data. It then validates these rules with the remaining 20 percent of the data that has not been used in the formation of the rules. This needs re-creation of dynamic rules every time this cycle is run, without insight into those few rules which are actually spoiling the overall performance of the recommendations.
[0012] Hence, looking at the problems in the prior art, there is a need of a system and method to measure the effectiveness and performance of commercial rules and learnt rules that recommend content to user (s). The system must ensure to understand the user (s) preference and fine-tune the rules. It must increase the user (s) engagement and enable user (s) to view the relevant content.
[0013] Summary of the invention:
[0014] The present invention overcomes the drawbacks in the prior art and provides a system to measure the effectiveness and performance of commercial rules and learnt rules that recommend content to user (s). In a preferred embodiment, the system comprises an input module and a display module installed in a mobile device, a setup module in a server and a data lake module in a data base module. The input module in the mobile device allows the user (s) to login and type the required content therein. The setup module in the server allows an admin to create a plurality of rules. The rules qualify the typed content by the user (s). The database module stores, processes and paginates the created rules to provide one or more recommendations based on the typed content thereby. The data lake module in the database module analyzes the recommended contents based on an intelligent decision making using the crated rules. Each recommendation is assigned and assembled with a rule identification therein. The display module displays the effective and relevant recommendations with their respective rule identification to the user (s) through the mobile device therein.
[0015] In a preferred embodiment, the system tracks and marks the interactions of the user (s) against the recommendations with their respective rule identification that are sent to the user (s) and used by the user (s).
[0016] In another embodiment, the method includes the steps of logging into an input module in a mobile device and allowing the user (s) to type the required content therein. After logging in the input module, a plurality of rules are created by an admin at a server. The rules are stored into a data base. After storing, the data in the server is assembled and the data is sent to user (s) at the input module, wherein every data point is assigned with a rule identification (ID) forming the recommendations. The recommendations are stored in a data lake. The recommendations are received by the user (s) and the user interacts with recommendations. The interactions are sent to the server with their rule IDs. The rules IDs are marked against the recommendations in the data lake. After marking the rule ID’s, the data is analyzed in the data lake and the performance of the rules is measured. Finally, the effective and relevant recommendations with their respective rule identification are displayed to the user (s) through the mobile device therein.
[0017] Brief description of the drawings:
[0018] The foregoing and other features of embodiments will become more apparent from the following detailed description of embodiments when read in conjunction with the accompanying drawings. In the drawings, like reference numerals refer to like elements.
[0019] FIGURE 1 illustrates the block diagrams of a system to measure the effectiveness and performance of commercial rules and learnt rules that recommend content to user (s), in accordance with an embodiment of the invention.
[0020] FIGURE 2 illustrates a process flow of a method to measure the effectiveness and performance of commercial rules and learnt rules that recommend content to user (s), in accordance with an embodiment of the invention.
[0021] Detailed description of the invention:
[0022] Reference will now be made in detail to the description of the present subject matter, one or more examples of which are shown in figures. Each example is provided to explain the subject matter and not a limitation. Various changes and modifications obvious to one skilled in the art to which the invention pertains are deemed to be within the spirit, scope and contemplation of the invention.
[0023] The present invention overcomes the drawbacks of the technology models available in the state of the art by providing a system to measure the effectiveness and performance of commercial rules and learnt rules that recommend content to user (s). The system comprises of a client module having a mobile application, a server module and a data base. A set of rules are created by admin of the system and recommendations are built using the rules. The system qualifies each recommended content with which rules resulted in the recommended content and list them against the content. The present system understands the effectiveness of the rules and more relevant content is provided to the user.
[0024] In the present invention the measurement of performance of rule helps in understanding the user’s preference and fine-tuning the rule accordingly. The present system eliminates the low performing rules and retains the rules which are performing well i.e. the rules which provide most relevant content to the user. The system increases the user (s) engagement and provides the relevant content to the user (s).
[0025] FIGURE 1 illustrates the block diagrams of a system to measure the effectiveness and performance of commercial rules and learnt rules that recommend content to user (s), in accordance with an embodiment of the invention. The system comprises an input module (102) and a display module (107) installed in a mobile device (102), a setup module (104) in a server (103) and a data lake module (106) in a data base module (105). The input module (102) in the mobile device (101) allows the user (s) to login and type the required content therein. The setup module (104) in the server (103) allows an admin to create a plurality of rules. The rules qualify the typed content by the user (s). The database module (105) stores, processes and paginates the created rules to provide one or more recommendations based on the typed content thereby. The data lake module (106) in the database module (105) analyzes the recommended contents based on an intelligent decision making using the crated rules. Each recommendation is assigned and assembled with a rule identification therein. The display module displays the effective and relevant recommendations with their respective rule identification to the user (s) through the mobile device therein.
[0026] In the preferred embodiment, the system qualifies each recommended content with which rules resulted in the recommended content and list them against the content. The system understands the effectiveness of the rules and more relevant content is provided to the user.
[0027] The user (s) interacts with the system, wherein all interactions are tracked and marked against the recommendations present in the data lake (105). The data lake (105) contains all the recommendations which are sent to the user and used by the user.
[0028] In the preferred embodiment of the invention, the data stored in the data lake (105) are analyzed for intelligent decision making. The analysis enables the measurement of the rules.
[0029] In the preferred embodiment of the invention, the mapping between the rule and recommendation, along with the context of recommendation are logged in the database (103). The recommended contents are pushed to the mobile application (104) in client’s module (101) with their respective rule identification (ID). Each user interaction in the client module (101) is measured and the user (s) sends back the data to the database (103) on which contents get accepted or interacted. The database (103) marks the performance of the rule (in terms of acceptance, interactions) against the logged recommendation (recommendation log). The data analysis in the database (103) is used to understand the effectiveness of the rules and more relevant content can be provided to the user. Higher the number of rules matching, higher the weight of the recommendation.
[0030] FIGURE 2 illustrates a process flow of a method to measure the effectiveness and performance of commercial rules and learnt rules that recommend content to user (s), in accordance with an embodiment of the invention. The method (200) includes the steps of logging into an input module in a mobile device and allowing the user (s) to type the required content therein. A plurality of rules are created by an admin at a server, at step (202). The rules are stored into a data base, at step (203). The data in the server is assembled by using the data stored in the database and the data is sent to user (s) at the client module, wherein every data point is assigned with a rule identification (ID) forming the recommendations, at step (204). The recommendations are stored in a data lake, at step (205). The recommendations are received by the user (s) and the user interacts with recommendations, wherein the interactions are sent to the server with their rule IDs, at step (206). The rules IDs are marked against the recommendations in the data lake, at step (207). The data is analyzed in the data lake and the performance of the rules is measured, at step (208). Finally, at step (209), the effective and relevant recommendations with their respective rule identification are displayed to the user (s) through the mobile device therein.
[0031] In the present invention, the measurement of performance of rule helps in understanding the user’s preference and fine-tuning the rule accordingly. The present system eliminates the low performing rules and retains the rules which are performing well. The present system understands the effectiveness of the rules and more relevant content is provided to the user.