Predicting Service Ratings using Geospatial Analysis of Social Media User Data

Abstract

This research investigates the correlation between the geographical location of social media users and their service ratings. The objective is to develop a predictive model leveraging geospatial data from platforms like Twitter and Yelp to forecast service ratings for businesses. We explore the impact of user location proximity to a service provider, neighborhood characteristics, and user demographic inferences from location data. The results demonstrate a significant improvement in rating prediction accuracy compared to existing methods, showcasing the potential of geospatial analysis for enhancing service quality assessments.

Introduction

Online service ratings significantly influence consumer choices and business reputation. Current rating systems rely heavily on user reviews, often neglecting valuable contextual information. This research addresses this gap by incorporating geospatial data – specifically, users' locations – to enhance the accuracy of service rating predictions. Understanding how location influences ratings can help businesses target improvements and consumers make informed decisions. Existing methods struggle to capture the spatial dependencies and neighborhood effects that impact service perception. This research aims to bridge this gap by developing a robust predictive model that integrates geospatial analysis with social media user data.

Objectives

• To develop a predictive model that accurately forecasts service ratings based on users' geographical locations and other relevant factors.
• To quantify the contribution of geospatial data to improved service rating prediction accuracy compared to existing methods.
• To identify geographical patterns and factors influencing service ratings.