Title - Banking domain Credit Card department’s data prediction
Scope: Risk analysis
Types of data - Structured, Semi-strutted data Machine Learning Algorithms: · Target column with Distributional values · Target column with binary values Technology stack – · Python · ML algorithms · SQLite (Structured source) · CSV, TSV, Multi-delimited values, JSON ( Semi-structured sources)
Project Duration:
Efforts in Weeks Tasks
1 Requirement analysis ,generate sample data ,
desired output - eligible or not
2 - python integration with source systems (SQLite, CSV, TSV, Multi-delimited, JSON) and data cleaning
1 Algorithm implementation and model training
1 Test algorithm results and check the error
N/A Git integration
1 Performance tuning
1 Documentation
CSV dataset description: Dataset contains 26052 records with 7 columns (fields) The description of columns (fields) : · City column has 986 distinct values · Date column has 20 months data · Card Type column contains 4 types of categories · Exp Type (expenditure type) column contains 6 types of values: Bills, Food, Entertainment, Grocery, Fuel, and Travel · Gender column · Amount column: contains transaction amount · EMI Paid: This column contains binary value about whether user paid EMI by due date or not Note: There are many other columns like transcation_id, card_number, etc. which are already filtered by the data source team.