Customer Decision Factors and Sentiment Analysis on Flipkart Product Review Data

Project Overview

Project Name: FACTORS AFFECTING CUSTOMER DECISION - SENTIMENTAL ANALYSIS USING FLIPKART PRODUCT REVIEW DATA

Institution: Dublin Business School, Ireland

Welcome to the repository for our research project, "Factors Affecting Customer Decision - Sentiment Analysis Using Flipkart Product Review Data." This project was conducted as part of the final semester research presentation at Dublin Business School, Ireland.

Abstract

This project addresses the need for a more accurate and reliable approach to determine product ratings based on customer sentiment in Flipkart product reviews. The primary objective is to predict product ratings based on customer reviews to enhance the accuracy of Flipkart's existing rating system.

Project Highlights

• Data Collection: We employed web scraping techniques to gather a comprehensive dataset of Flipkart product reviews. The data collection code is available in the data_collection folder.

• Datasets: The datasets folder contains seven different datasets, each representing product reviews extracted from Flipkart.

• Analysis and Evaluation: Our primary analysis, data cleaning, sentiment analysis, rating prediction, and evaluation are conducted in the analysis_and_evaluation folder. This code file encompasses the entire analysis, including data preprocessing and performance evaluation.

• Project Report: For detailed insights into the analysis and evaluation, you can access our comprehensive project report in the Project_Report folder. The report contains in-depth explanations of the methodology and results.

• Documentation: Additional project documentation, resources, and presentations can be found in the 'Documentation' section.

Getting Started

To replicate our research and analysis, follow the instructions provided in the respective project folders. Detailed documentation and resources are available in the 'Documentation' section.

Data Collection

1. Navigate to the data_collection folder to find the Python script and/or notebook used for web scraping Flipkart reviews.

Analysis, Data Cleaning, and Evaluation

1. The main analysis, including data cleaning, sentiment analysis, rating prediction, and evaluation, is conducted in the analysis_and_evaluation folder. Start with the notebook file titled Ajumon_Remesan_Project_Review_Analysis.ipynb. This notebook contains the merging of datasets, data cleaning, sentiment analysis, rating prediction, and graphical representations of the analysis results.

Project Report

1. Access the detailed project report for a comprehensive understanding of the analysis and evaluation here.

Conclusion

Our project demonstrates the utility of sentiment analysis in predicting customer ratings based on their reviews. By implementing this approach, businesses can obtain more accurate and reliable feedback, leading to improved customer satisfaction and informed decision-making processes.

For any questions or collaborations, feel free to contact the project contributors. Thank you for your interest in our research project!