Mining Online Product Evaluation System based on Ratings and Review Comments

ABSTRACT:

In the digital era, online product evaluation systems play a crucial role in guiding customers’ purchasing decisions by providing insights into product quality and customer satisfaction. With the vast expansion of e-commerce platforms, consumers heavily rely on ratings and review comments to make informed decisions. However, the credibility and effectiveness of these evaluations depend on efficient data processing and analysis. To address this, we propose a Mining Online Product Evaluation System, which systematically collects, analyzes, and visualizes product reviews and ratings to enhance user experience and decision-making.

The need for such a system arises due to the increasing volume of online shopping transactions and the vast number of product reviews generated by customers. Manual interpretation of these reviews is impractical, making automated review analysis a necessity. Our system not only provides a platform for users to rate and review products but also incorporates sentiment analysis to categorize reviews as Very Positive, Positive, Neutral, Negative, or Very Negative. Additionally, it generates graphical insights to help administrators and sellers understand user preferences and improve product offerings accordingly.

This system is built using Java for backend development, JSP, CSS, and JavaScript for the frontend, and MySQL as the database. The integration of automated sentiment analysis, graphical data representation, and an interactive user interface makes this platform an efficient solution for mining and evaluating online product reviews. By providing insights into customer sentiments, it helps businesses refine their products and services while enhancing user trust and satisfaction in the online marketplace.

PROJECT OUTPUT VIDEO:

EXISTING SYSTEM:

The existing system for online product evaluation primarily relied on basic rating mechanisms and textual review submissions. These systems allowed users to rate products using a numerical scale, typically from 1 to 5 stars, and write review comments based on their experiences. The ratings were aggregated to provide an overall score for each product, helping new buyers get a general idea of product quality.
In the existing system, in addition to ratings, customers could post written reviews, which were displayed alongside the products. Users could manually browse through these reviews to understand different opinions about a product. Some systems also featured simple filtering options, enabling customers to sort reviews based on date, relevance, or rating scores (e.g., highest-rated or lowest-rated).
The existing systems also provided basic administrative controls, where administrators could manage product listings, update prices, and monitor user reviews. Some platforms allowed product owners or administrators to respond to customer reviews, offering clarifications or addressing concerns raised by buyers. Users could also flag inappropriate reviews, ensuring that irrelevant or misleading content was moderated.
Overall, the existing systems provided a structured way for customers to evaluate products based on shared experiences. They laid the foundation for modern online review analysis by offering essential functionalities such as rating aggregation, review posting, and product management.

DISADVANTAGES OF EXISTING SYSTEM:

Despite offering fundamental functionalities for product evaluation, the existing systems had several limitations that affected their efficiency and reliability.
Lack of Automated Sentiment Analysis – The previous systems relied solely on numerical ratings and textual reviews without analyzing the sentiment behind user comments. This made it difficult to extract meaningful insights from reviews, as users had to manually read through multiple comments to understand customer opinions.
No Graphical Representation of Product Ratings – The absence of graphical analysis meant that users had to rely only on average ratings without a visual comparison of how different products performed. Without pie charts or other dynamic graphs, analyzing customer feedback patterns was time-consuming.
Limited Review Filtering and Categorization – The earlier systems provided only basic sorting options but lacked advanced filtering capabilities. Users could not filter reviews based on sentiment (e.g., positive, neutral, negative), making it difficult to focus on specific feedback categories.
Lack of Real-time Insights for Admins – Administrators had no efficient way to analyze user feedback trends dynamically. They had to manually go through reviews to understand user preferences and product performance, which was time-consuming and inefficient.
No Precision-based Review Analysis – The system did not provide detailed precision metrics for sentiment analysis. Users could not see an in-depth breakdown of how many reviews were classified as very positive, positive, neutral, negative, or very negative.
Limited User Engagement – Without interactive and automated review insights, user engagement was restricted. Customers had to rely on static information without any AI-powered suggestions or insights to help them make informed decisions.
These limitations highlighted the need for an improved system with advanced sentiment analysis, real-time graphical insights, and better review filtering mechanisms to enhance the overall user experience and decision-making process.

PROPOSED SYSTEM:

The proposed Mining Online Product Evaluation System based on Ratings and Review Comments is designed to provide a more advanced and insightful approach to analyzing product reviews and ratings. This system is developed using Java as the coding language, with JSP, CSS, and JavaScript for the frontend and MySQL as the database. The platform enables users to evaluate products effectively while allowing administrators to manage products and analyze customer feedback using automated sentiment analysis and graphical representation.
The developed system consists of two main entities: User and Admin. The Admin panel provides functionalities such as adding and managing products, viewing users, monitoring ratings and reviews, and performing graph-based analysis. Admins can add new products by entering details such as product name, description, price, category (MOBILE, LAPTOP, ACCESSORIES, OTHERS), tags, and product images. They can view all registered users along with their personal details and access comprehensive user feedback on products. Through Graph Analysis, administrators can visualize product performance based on user ratings using dynamic pie charts displaying average ratings for each product.
On the User side, individuals must first register by providing essential details such as name, email, date of birth, gender, phone number, address, and password. Once logged in, users can search for products by entering the product name and selecting a category. If a product is available, users can view its detailed information, including images, descriptions, pricing, and tags. The system allows users to submit reviews and ratings (1 to 5 stars), which are then processed for sentiment analysis. The system determines the overall sentiment of a review based on natural language processing (NLP), categorizing it into predefined sentiment levels. Additionally, precision values for each sentiment category are displayed, ensuring transparency in the sentiment analysis results.

ADVANTAGES OF PROPOSED SYSTEM:

The Mining Online Product Evaluation System based on Ratings and Review Comments offers numerous advantages over traditional review management systems. By leveraging advanced technologies and sentiment analysis, the proposed system enhances the efficiency, accuracy, and user experience of product evaluation.
Improved Product Evaluation and Insights: The system integrates automated sentiment analysis, enabling users and administrators to gain deeper insights into customer reviews. This allows businesses to understand customer preferences and improve product offerings based on real-time feedback.
Enhanced User Experience: Users can easily search for products using product names and categories, making it more convenient to find relevant items. Additionally, the intuitive JSP-based frontend with CSS and JavaScript ensures a smooth and interactive browsing experience.
Efficient Product and Review Management: The admin panel provides a structured approach to managing products, users, and reviews. Graph Analysis presents a visual representation of product ratings, helping administrators identify top-performing products and those requiring improvements.
Sentiment-Based Review Analysis: Unlike traditional systems that display reviews without classification, the proposed system automatically analyzes user comments and categorizes them into Very Positive, Positive, Neutral, Negative, or Very Negative sentiments. Additionally, precision values offer a more detailed breakdown of sentiment trends.
Secure and Organized User Management: With user registration and authentication, the system ensures that only registered users can provide reviews. This helps maintain data integrity and prevents spam or fake reviews, improving the credibility of product evaluations.
Dynamic Graphical Visualization: The Graph Analysis module dynamically generates pie charts to visually represent average ratings per product, making it easier for administrators to track product performance over time.
Scalability and Flexibility: The system is designed to handle multiple categories like MOBILE, LAPTOP, ACCESSORIES, and OTHERS, with the possibility of adding more categories in the future. The MySQL database efficiently manages large volumes of product data, user reviews, and ratings.
Real-Time Feedback System: Users can instantly submit reviews and ratings, and the system processes and analyzes them in real time. This ensures that administrators have up-to-date insights into customer satisfaction and product performance.
Accurate Rating System: The structured 1 to 5-star rating system combined with sentiment analysis ensures accurate and meaningful product evaluations, helping future customers make informed purchasing decisions.
Better Decision-Making for Businesses: With access to sentiment-based review analysis and graphical reports, businesses can make data-driven decisions regarding product improvements, pricing strategies, and customer engagement.
By integrating advanced sentiment analysis, graphical reporting, and a user-friendly interface, the proposed system enhances the overall efficiency of online product evaluation, making it more reliable, insightful, and user-centric.

SYSTEM REQUIREMENTS:

HARDWARE REQUIREMENTS:

System : Pentium i3 Processor.
Hard Disk : 20 GB.
Monitor : 15’’ LED.
Input Devices : Keyboard, Mouse.
Ram : 4 GB.

SOFTWARE REQUIREMENTS:

Operating system : Windows 10/11.
Coding Language : JAVA.
Frontend : JSP, CSS, JavaScript.
JDK Version : JDK 23.0.1.
IDE Tool : Apache Netbeans IDE 24.
Tomcat Server Version : Apache Tomcat 9.0.84
Database : MYSQL.

Java Projects

Mining Online Product Evaluation System based on Ratings and Review Comments