Decode transaction data information
E-commerce allows consumers to electronically exchange goods and services with no barriers of time or distance. Electronic commerce has expanded rapidly and is predicted to continue at this rate, or even accelerate. In the near future, the boundaries between “conventional” and “electronic” commerce will become increasingly blurred as more and more businesses move sections of their operations onto the Internet. Most e-commerce companies have no idea how to utilize their tons of transaction data. Our project aims to build a data analysis model to produce a business report that can explain the information hidden in the transaction data in “plain” language.
E-commerce companies need to upload their transaction data onto our platform. For each transaction, we may need the order number, customer information, order date, product, sales revenue, and quantity. The order date can help us to do time series forecasting of the future sales, profit and quantity. The statistical analysis provides top selling products by category and by region. The collaborative filtering predicts which product customers are more likely to buy. The Apriori Algorithm suggests a recommendation system for bundle sales.