Bigmart Sales Prediction

The data scientists at BigMart have collected 2013 sales data for 1559 products across 10 stores in different cities. Also, certain attributes of each product and store have been defined. The aim is to build a predictive model and predict the sales of each product at a particular outlet.

Using this model, BigMart will try to understand the properties of products and outlets which play a key role in increasing sales.

Given the problem, three solutions were developed:

• Complete data analysis using descriptive and statistical techniques such as analysis of histograms, boxplots, bar graphs, lines, analysis of missing values and outliers and data transformation;
• Machine Learning: Predictive modeling testing different machine learning algorithms such as: Decision Tree, RandomForest, XGBoost, LightGBM and SVM;
• Deep Learning: Predictive modeling using deep neural networks with Tensorflow and Keras.