Machine Learning based Approach for Predicting Future Purchases in E-Commerce
Ayaz Khan, Department of Data Science, University of Engineering and Technology, Peshawar, Pakistan.
Muhammad Aaqib, Department of Mechatronics Engineering, University of Engineering and Technology, Peshawar, Pakistan.
Corresponding Author:
Ayaz Khan (ayaz.khan882@gmail.com)
Abstract:
Now-a-days, numerous individuals use e-commerce platforms to purchase their daily essentials. E-commerce is now an integral part of how people buy things. However, as the e-commerce industry expands, so do the associated challenges, which are also growing. The motivation behind this research is to better understand the domain of e-commerce by investigating publicly available datasets. This section focuses on the different stages involved in the selection and classification of product categories. It's not an easy task, as there are many factors that can determine whether a product will be successful or not. Making the wrong decision can have a huge impact on the outcome, either leading to significant success or failure. Therefore, e-commerce data analysis is used to predict changes in customer behavior based on various factors such as product quality, categories, brands, purchase time cycles, and purchase weekdays. The proposed study employed various machine learning classifiers such as random forest, K-nearest neighbor, logistic regression, and gradient boost. We predicted and identified from the simulation process some useful outcomes to help sellers anticipate potential buying trends and take advantage of them while maintaining sales growth.
Keywords:
Machine Learning; E-Commerce; Sale Forecast; Product Classification