The objective is to improve online retail experience of a user by gaining insights into user behaviour and recommending the user’s next purchase.
In this project, we try to build an web application with a simple UI that work on a very typical and classical recommendation system scenario, that is product recommendation in E-commerce, i.e., Amazon.
The data required for this project is taken from the below website.
It contains user reviews (numerical rating and textual comment) towards amazon products on 24 product categories(e.g., cell phones, clothing, beauty, etc.), and there is an independent dataset for each product category. We will select 5 product categories in this project i.e., Arts.txt.gz, Cell_Phones_&_Accessories.txt.gz, Jewelry.txt.gz, Musical_Instruments.txt.gz, Watches.txt.gz. On choosing the category of product, recommendations are displayed based on user based similarity .
After we select a dataset to work on, this project will mainly consist three steps:
- Data Processing
- Perform EDA
- create the training and testing datasets
- Conduct rating prediction and make evaluation
- Conduct Top-N Recommendation