International Journal for Research and Development in Engineering (IJRDE) www.ijrde.com Vol.1: Issue.

2, October-November 2012 pp- 37-45

DATA MODELING BASED ON USAGE ANALYTICS IN MAINFRAME CICS APPLICATIONS

1

R.Krishna Kumar, 2G.Radhika, 3 M.Manikandan, 4Abhinav Mishra, 5YogeshKumar Patel, 6 Priya Bose 1 2 ( krishnakumar.r2@tcs.com, g.radhika1@tcs.com, 3 manikandan.m4@tcs.com, 6 4abhinav.mishra2@tcs.com, 5 yogeshkumar.patel@tcs.com , priya.bose2@tcs.com)
TATA CONSULTANCY SERVICES, Trivandrum, India

Abstract
If you query the phrase 'Preference Data Choice' on the Internet, you may find a range of text focusing on technology frameworks to support customer decision, software to manipulate data and professional 'quick tips' to attract customers. Internet marketing means an unrestricted opportunity to inundate the customers with comprehensive product information. It gives the customer a sense of choice over product selection they view on a website can result in a more positive view of the company, thereby making them more likely to buy the product. A Human-Computer Interaction should devote more attention in helping the users to make better preferential choices. The more you involve the customers in selecting an item from a category of same type, the more they grow attached to the company. It goes to an extent that they become resistant to negative information about the product. Keywords: Preferential data choice, TSQ, XCTL, CICS.

1. INTRODUCTION
Many Human-Computer Interaction design principles, guidelines, and practices can be seen as supporting both preferential and nonpreferential choice; but preferential choice gives rise to characteristic cognitive processes that deserve special attention. The preferential choice can be given in two ways: I) Based on psychological research that is not specific to any domain. II) Or displaying a few common things based on the different attempts by the users. Websites that do not offer a choice of product for customers may not be preferred much. Preferential choice can contrast with non preferential choice, which is a matter of choosing the single correct action for achieving a given goal. Using the customer's previous choice of interest, builds richer understanding of which product he is interested in. Now a days most firms practice this target marketing, to tailor different offers to different customers depending on their previous purchase history.

In our sample application we had two types of users 1. Guest 2. Registered user
37

International Journal for Research and Development in Engineering (IJRDE) www.ijrde.com and an Administrator. The Guest can only View our products. Only a registered user can buy the products. Once the user's purchase is made, the details of the user are recorded into the database. Upon user's subsequent visit to our application, he will be provided with a suggestion option which suggests him a list of products based on his previous purchase. From the list of products available in the suggestion option, the user gets the privilege of buying the items that is his cup of tea. In our application, we have two main categories of products one being Books and the other being Electronics. ALGORITHMS 2.1 Preferential Data Choice: The algorithm for Preferential Data choice is as follows: Step I: Get the Customer ID (Unique for Each Customer). A unique Customer ID is allotted to each user after registration. This customer ID is used for identifying the history of products purchased by the user. Step II: Identify the Products available Both Product code and Customer ID are used as a condition for selecting the Vol.1: Issue.2, October-November 2012 pp- 37-45 available products for that user. Step III: Group each Product under one category and find the Total Once the products are identified they are grouped under similar category and their sum based on quantity is calculated. Step IV: Compare the Two main categories and determine the larger Quantity. Upon finding the total for each category, the comparison is made on the basis of their total purchase. Step V: Depending on the Quantity, display the customers suggestion. As a result of the comparison, a list of suggestion is given to the user. The Customer purchase history is stored in the Track_Details Table. The Quantity is retrieved from this table using the Customer ID and stored in Working-Storage variable. The figures below show how it is implemented in Mainframes.

38

International Journal for Research and Development in Engineering (IJRDE) www.ijrde.com Vol.1: Issue.2, October-November 2012 pp- 37-45

Figure 2.1.1. Code for retrieving the Quantity for Books category

Figure 2.1.2 Code for retrieving the Quantity for Laptop (Electronics) category

29

International Journal for Research and Development in Engineering (IJRDE) www.ijrde.com Vol.1: Issue.2, October-November 2012 pp- 37-45

Figure 2.1.3. Code for retrieving the Quantity for PC (Electronics) category

Figure 2.1.4. Code for retrieving the Quantity for Mobile (Electronics) category

30

International Journal for Research and Development in Engineering (IJRDE) www.ijrde.com Vol.1: Issue.2, October-November 2012 pp- 37-45

Figure 2.1.5. Comparison of the sum of quantities of all the books and electronics purchased by a User

31

International Journal for Research and Development in Engineering (IJRDE) www.ijrde.com Based on the comparison, corresponding suggestion list is given to the user. These suggestions are created using the Map concept in CICS, in which each Map lists items based on his purchase history. Each Suggestion Map is called using a XCTL calling command in CICS which includes all his products. Dynamic Graph: In our application we have developed a graph that varies dynamically based on the frequency of a product purchased. The algorithm for the same is as follows: Step I: Calculate the sum of quantities on the basis of their product codes. Sum of quantities are calculated based on the different products. Vol.1: Issue.2, October-November 2012 pp- 37-45 Step II: Obtain the total for each subcategory. Each sum is placed under the respective sub-category and the total for each sub-category is calculated Step III: Calculate the percentile. The percentile of each product within a particular sub-category is calculated. Step IV: In order to plot the graph, total is scaled to 10. In order to plot the graph, the percentile is scaled to 10 and the values are shown in the form of bars. Step V: Press any key to obtain the graph. The X-axis represents the categories and Yaxis represents the quantities.

Figure 2.2.1

42

International Journal for Research and Development in Engineering (IJRDE) www.ijrde.com Vol.1: Issue.2, October-November 2012 pp- 37-45

3. FLOW CHART A flow chart describing our algorithm:

5. CONCLUSION 4. APPLICATION OF PREFERENCE DATA CHOICE Application of Preference Data Choice is very vast. It is mainly used in Online Shopping Websites, Airlines and Tourism related applications. Preference Data Choice concept is mainly implemented to attract Customer attention and the frame of reference explicit to the customers via the inclusion of an array of attribute and product alternatives helps in maintaining a long lasting relationship with them. It is very evident that implementing Preferential Data Choice in an online application will attract the customers' attention and in-turn increases the revenue of the organization. The user feels that his preferences are being given importance and thus he becomes a repeat customer. The attitude and intention of a customer towards an application are not only affected by the ease of use, usefulness, and enjoyment, but also by exogenous factors like consumer traits, situational factors, product characteristics, previous experiences, and trust in an application.

44

International Journal for Research and Development in Engineering (IJRDE) www.ijrde.com Vol.1: Issue.2, October-November 2012 pp- 37-45
Journal of Transportation Economics and Policy, 30, 203212. 6. REFERENCES [1] C Choudhury, RAND Europe, UK and Massachusetts Institute of Technology, US Dealing with Repeated Choices in Stated Preference: An Empirical Analysis, European Transport Conference 2008. [2] Ouwersloot, H. and P. Rietveld (1996) Stated choice experiments with repeated observations, [3] Designing application. and programming CICS

[4] CICS transaction server for z/OS, Version 4.2 IBM. [5] Designing application. and programming CICS

45

International Journal for Research and Development in Engineering (IJRDE) www.ijrde.com Vol.1: Issue.2, October-November 2012 pp- 37-45

APPENDIX

Figure A- This is the Suggestion map that will be displayed when a customer prefers Electronics over Books.

45