Abstract:
Modern vending machines offer a wider range of products and services, including fresh food, electronics, beauty products, and even rental services like bicycles or umbrellas. they provide convenient access to goods and services in locations where traditional stores might not be feasible or available. I got curious about this and decided to dive into vending machine data. I wanted to see them from both the customer's view and the perspective of the people who manage them, using data to guide my investigation. I gather data from vending machines across different locations. The locations include a library, a mall, an office location, and manufacturing locations. I have used Regression (XGBRegressor) analysis to predict sales volume based on factors such as product price, location demographics, and time of day. I also used k-Nearest Neighbors (kNN) for analyzing mapped quantity sold in vending machine data involves predicting the quantity of products sold based on the characteristics of neighboring data points. Additionally, I also used Linear Regression which is employed to identify linear relationships between sales and various influencing factors, while a Decision Tree provided a visual representation of decisions, and their possible outcomes based on different variables affecting sales. Assess metrics such as accuracy, precision, recall, and R-squared value to measure the model's effectiveness in predicting vending machine behavior. According to the conclusions of this research, Vending machine sales thrive on technology, diverse offerings, and easy payments. Meeting consumer preferences and managing inventory efficiently is key to success.
