Big Mart Sales Data Visualization and Correlation

Abstract

The amount of unprocessed data available every day is growing. This massive amount of data needs to be effectively assessed to give results that are extremely useful. In the present day, it is crucial for inventory management and demand forecasting to collect sales data for commodities or things, together with all their numerous dependent or independent parts. In a Big Mart Company, the use of sales forecasting is to estimate numerous goods that are readily available and supplied at multiple retailers in different towns. As the number of products and outlets increased drastically, it became increasingly difficult to forecast them manually. As a result, it is necessary to see to what extent the relationship between several variables, including price, popularity, time of day, outlet type, outlet location, etc., affects the appeal of a product. In this research, a data cleaning process was carried out, and data visualization using scatter plots, as well as finding Pearson correlations. The raw processing the data with study of a case big mart sales data is taken from the Kaggle website [https://www.kaggle.com/datasets/sandeepgauti/bigmart-sales]. The Pearson correlation test determines a lack of connection between the two Item_Weight and Item_Outlet_Sales variables. There is a strong but negative correlation where if Item_Visibility decreases, Item_Outlet_Sales also decreases. Positive relationships exist between the two Item_MRP and Item_Outlet_Sales variables. In addition to the correlation test, descriptive statistical analysis is also performed here. With this simple data processing, the raw data will be better organized and easier to analyze, read, and use.