Business Response Prediction based on Consumer Behaviour Patterns

This survey paper presents a comprehensive analysis of predictive models for business response in the hospitality industry, focusing on cafes and restaurants. Leveraging data from Zomato, a popular restaurant aggregator platform, and supplemented with location data, we explore effective strategies for preprocessing and clustering based on key parameters, including location, cuisine, nearby institutions, parking, traffic, bestsellers, cost, ratings, delivery options, and establishment type (cloud kitchen/dine out). We investigate a variety of machine learning and statistical approaches tailored to distinct clusters of cafes and restaurants. Our evaluation involves training and testing models to predict business responses, considering varying customer foot traffic and preferences. The findings emphasize the importance of precise clustering and customized predictive models for improved accuracy in response forecasting. The insights derived from this survey are anticipated to significantly impact decision-making processes for businesses in the hospitality sector.