Course Hero Logo

Yamini Final.pdf - Credit Card approval Prediction using...

Course Hero uses AI to attempt to automatically extract content from documents to surface to you and others so you can study better, e.g., in search results, to enrich docs, and more. This preview shows page 1 - 2 out of 3 pages.

Credit Card approval Prediction using MachineLearningYamini Vakula Madhavi Nannapanenicomputer ScienceLewis universityIllinois, United states[email protected]Abstract—By using Machine Learning (ML) and data analyticsin banking industry is become a trend in the market to competewith others to reduce credit risks. As due to huge volumes ofdata that is generated through different banking organizationsand business transactions can pose a great challenge of com-putational and storage for all data analysis and smart tasks.In order to address the tasks associated with smart tasks, cloudcomputing (CC) is evolved. In this paper we will be proposing andimplementing a ML based technique for all credit cards to knowtheir credit scores using financial risk controls and compare itsperformance with traditional scoring system. As the traditionalcredit card scoring system, is one of the complicated methodsthat has restrictions on selection of specified variables that haslimitations. By implementing such methods the internet age datainstructions can be limited which means it cannot analyse thedata that has high dimension, complexity and non-linearity.Index Terms—component, formatting, style, styling, insertI. INTRODUCTIONCredit Worthiness (CW) is the validation process that isdone by lender who tries to determine if there is any possibilityof borrower to default on his debt obligations. This is referredas the credit score by financial control of the organizations.The CW of the company or individual is majorly based onCredit rating systems. The higher the credit score then highCW. Payment history , credit scores, health status and howthe person meets his debt obligations generates the CW of theperson. In general 35II. THE PROBLEM:Unbalanced data and in accurate data by prediction are themajor problems associated with this research. There is alwaysa unbalanced data that is not normalized due to heavy loadof data is pushed. Due to this forceful data, there is will be aunbalance in the retrieval of client information when he inputshis information to check his credit score. Prediction of datacannot be always right as it sometimes predicts in accuratedata especially when user gives his in correct details. Bankingorganizations transacts large volumes of data every day as apart of their business transactions so here there might be achance of data breaches or loss of data to the end user. Thisloss of data creates more space to add non predictable datathat is found using the predictable classifier. Surveying of thenew machine learning models for prediction of the user creditscore can be more easy when considered with all challengesthat are likely to be triggered in the market. The credit scoreresults can be good or bad depends on banks approval on thecredit scoring system. It is generally based on the paymenthistory, background etc. Sometimes there might be a chanceof fraud on the credit card as if someone know the Social

Upload your study docs or become a

Course Hero member to access this document

Upload your study docs or become a

Course Hero member to access this document

End of preview. Want to read all 3 pages?

Upload your study docs or become a

Course Hero member to access this document

Term
Summer
Professor
N/A
Tags
credit score, Credit history, Credit rating

Newly uploaded documents

Show More

Newly uploaded documents

Show More

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture