Developing a machine learning model for predicting purchasing behavior of F-commerce-based customer

Abstract

The modern tech-based lifestyle has become much easier with online purchases. Social media platforms like Facebook, Instagram and Twitter have taken it to another level. Online business through Facebook is now a significant topic worldwide, so are the challenges. With the easy access of internet and free use of Facebook, doing business through Facebook platform is not a big deal. But as anyone, anywhere can start a small business with little capital using this giant platform due to its easiness, the competition and long-time survival is also the biggest of all challenges. Hence the aim of this study is to build a model that would help anyone get acknowledged with the satisfaction level of F-commerce users and target those pleased consumers in the future to serve better and minimize the survival challenges. The objective of this project is to predict the F-commerce based customer’s satisfactory level and next preferable purchase. In this study, a survey method has been used to get real time customer data and a machine-learning model has been made that would help to cluster those customers together who prefer to purchase similar categories of product from Facebook. This project then also aims to classify and predict the satisfaction level of those customers in their purchase behaviour with real-world data by using K-medoid with the PCA (Principal Component Analysis) algorithm and SVM (Support Vector Machine).