ME-781(Statistical Machine Learning and Data Mining) Course Project | Prof. Asim Tewari | IIT Bombay
Millennials have become a major part of our customer base now and as the internet reach is widening into all generations, every person with a mobile is a potential customer for e-commerce sites. This shift makes predicting customer behavior all the more pertinent and gives you an edge in the competition.
Once we help you predict customer behavior and customize their shopping process it will help boost sales, increase customer satisfaction and will certainly result in higher conversion rates and competitive advantage.
- We took a “customer_shoppers_intentions.csv” and extracted independent features which affect the Odds of customer Purchase.
- We Did Descriptive Data Analysis on the data set to understand the data and trends into the data.
- Data Processing was done to convert data from categorical to numerical keeping in mind ordinal, nominal data types.
- The dataset is divided into 80:20 train-test ratio and these algorithms are compared based on their accuracy on the test dataset.
- Classification and Clustering was used since outcome is binary.
- We Used Different Classifier to know which one is working best in aforementioned conditions.We Used all the Theories (Models) Taught In Class relevant to classification task.
- from our method we found that Neural Network Classifier is working best in our case