Online Movie Recommendation System

ABSTRACT:

The rapid increase in the number of online streaming platforms has led to an exponential increase in available movies, making it difficult for users to select content that aligns with their preferences. To address this challenge, the Online Movie Recommendation System is developed as an intelligent platform that provides personalized movie recommendations based on user ratings.

The Online Movie Recommendation System is developed using Java as the coding language, with JSP, CSS, and JavaScript for the frontend, and MySQL for the database. This system ensures that users receive movie suggestions tailored to their latest preferences, rather than offering generic recommendations. The system also incorporates sentiment analysis on a movie dataset comprising approximately 4,000 records, including movie IDs, titles, and critics’ consensus, to further enhance the recommendation process.

The need for such a recommendation system arises from the growing demand for personalized content delivery. Traditional recommendation methods often fail to adapt dynamically to users’ changing tastes. The proposed system overcomes this limitation by continuously analyzing user ratings and updating recommendations accordingly. Unlike conventional systems where all users receive similar recommendations, this platform ensures that recommendations are unique to each user and evolve based on their latest ratings. Additionally, sentiment analysis of critics’ consensus helps in providing insights into audience perceptions, enhancing user engagement and decision-making.

Overall, the Online Movie Recommendation System enhances the movie discovery experience by offering personalized recommendations, leveraging sentiment analysis, and providing an intuitive user interface. This project effectively bridges the gap between overwhelming content availability and user preferences, ensuring a seamless and engaging movie selection process.

PROJECT OUTPUT VIDEO:

EXISTING SYSTEM:

In the existing movie recommendation systems, suggestions were primarily based on predefined algorithms that categorized movies according to genres, release years, or popularity trends. These systems generally followed a rule-based approach, where movies were recommended based on fixed attributes such as top-rated films, most-watched content, or manually curated lists. Users typically received recommendations that were either static or limited to popular selections, without personalization based on their individual preferences.
In the existing recommendation systems used content-based filtering, where movies were suggested based on similarities in metadata, such as genre, director, actors, or storyline. This approach relied on structured movie information but did not consider user feedback dynamically. Other systems implemented collaborative filtering, which suggested movies by analyzing user behavior and preferences. In this approach, if multiple users had similar movie-watching patterns, they were recommended the same movies. However, these recommendations were largely influenced by general viewing habits rather than a user’s latest preferences.
In addition to algorithm-based recommendations, existing systems relied on manual process, where experts or platform administrators selected and categorized movies into different recommendation lists, such as “Trending Now,” “Editor’s Picks,” or “Top Movies of the Month.” These lists provided a broad selection of films but did not adapt to individual user choices dynamically.
Furthermore, sentiment analysis was not commonly integrated into existing recommendation systems. The impact of critics’ opinions and audience reviews was often presented separately from movie recommendations, limiting their role in the decision-making process. While reviews and ratings were available, they were primarily used as reference points rather than being actively incorporated into the recommendation mechanism.
Overall, the existing movie recommendation systems provided structured yet generalized recommendations, primarily focusing on genre-based classification, collaborative filtering, or expert-curated lists. While these approaches helped users discover movies, they lacked the ability to adapt dynamically to users’ evolving preferences and recent rating patterns.

DISADVANTAGES OF EXISTING SYSTEM:

While existing movie recommendation systems provided structured suggestions, they had several limitations that affected their effectiveness in delivering personalized recommendations. Some of the key disadvantages include:

Lack of Personalization: Traditional recommendation systems often relied on static algorithms, such as popularity-based or genre-based filtering. As a result, all users received similar recommendations, regardless of their unique preferences. These systems did not adapt dynamically to individual users’ changing tastes.
Limited Consideration of User Ratings: The existing systems either did not incorporate user ratings or relied on historical ratings without updating recommendations in real time. This meant that users could not receive new suggestions based on their most recent preferences, leading to outdated or irrelevant recommendations.
Dependence on Fixed Attributes: Content-based filtering approaches focused on attributes such as genre, director, or cast, limiting recommendations to movies with similar metadata. This approach failed to consider diverse user interests and did not introduce users to movies outside their immediate preferences.
Challenges in Collaborative Filtering: Collaborative filtering, which recommended movies based on similar user preferences, faced challenges when dealing with new users (cold start problem) or movies with limited ratings. If a movie had fewer ratings, it was often not recommended, even if it aligned with a user’s interests.
Lack of Sentiment Analysis Integration: The existing systems did not incorporate sentiment analysis of critics’ consensus or audience reviews. This meant that users had to rely on general ratings rather than understanding the sentiment behind them. A movie with an average rating might have mixed or polarized reviews, but users had no way of assessing these nuances.
Static Recommendation Mechanism: The existing systems did not continuously update recommendations based on a user’s latest ratings. Once a user received a recommendation, it often remained unchanged, even if their preferences evolved over time. This resulted in repetitive suggestions, reducing user engagement.
No Graphical Representation of Data Insights: Many earlier systems lacked visual analytics to help users and administrators understand movie trends. There were no dynamic graphs to display rating distributions, genre preferences, or user engagement trends, making data interpretation less intuitive.

Due to these limitations, traditional recommendation systems struggled to provide highly accurate and user-centric suggestions. As a result, users often had to browse manually or rely on external reviews to find movies that truly matched their preferences.

PROPOSED SYSTEM:

The Online Movie Recommendation System is designed to provide a dynamic and personalized movie suggestion experience based on user preferences and sentiment analysis. Developed using Java, with JSP, CSS, and JavaScript for the frontend and MySQL as the database, this system ensures that each user receives recommendations that adapt in real-time based on their latest ratings. Unlike earlier systems that relied on static filtering methods, the proposed system leverages a personalized approach where recommendations are updated dynamically as users provide new ratings.
The system consists of two primary entities: Admin and User. The Admin module provides functionalities such as adding movies with details like language, movie name, star cast, genres, and movie poster images. Admins can also manage a dataset of 4,000 movie records, view user ratings, and analyze trends through dynamic graph visualizations that display rating distributions across different genres. Additionally, the admin can perform dataset management tasks, including uploading, viewing, and deleting existing movie data.
The User module allows new users to register and log in to access various features. Users can browse through the list of available movies, provide ratings (on a scale of 1-5), and receive movie recommendations based on their most recent ratings. Unlike traditional recommendation systems where all users receive similar suggestions, this system ensures that each user gets a unique set of recommendations tailored to their evolving preferences. Additionally, sentiment analysis is integrated into the system, enabling users to analyze movie sentiment based on critics’ consensus. By selecting a movie from a dropdown list, users can view whether critics’ opinions are predominantly positive or negative.
The system also provides a graphical representation of movie ratings, helping the admin analyze trends based on genre preferences. This feature ensures a more insightful understanding of user interactions and movie popularity. The personalized recommendation engine continuously updates suggestions, ensuring that users receive fresh recommendations each time they interact with the platform.
With its dynamic recommendation algorithm, real-time rating updates, and sentiment analysis features, the proposed system enhances the movie discovery experience by making it more user-centric, interactive, and insightful.

ADVANTAGES OF PROPOSED SYSTEM:

The Online Movie Recommendation System offers several advantages over traditional recommendation models, making it more efficient, personalized, and user-friendly. The key benefits of the proposed system include:
Personalized Movie Recommendations: Unlike existing systems that provided static or generalized suggestions, our proposed system dynamically updates recommendations based on each user’s latest ratings. Every user receives unique suggestions tailored to their most recent preferences.
Real-Time Adaptive Recommendations: The system continuously refines its movie suggestions whenever a user rates a film. This ensures that recommendations evolve with changing user preferences, improving accuracy and relevance.
Integration of Sentiment Analysis: A significant enhancement in the proposed system is the use of sentiment analysis on critics’ consensus. Users can select a movie and analyze whether the overall sentiment is positive or negative, helping them make informed choices beyond just numerical ratings.
User-Friendly Interface: The system features a well-structured and intuitive interface using JSP, CSS, and JavaScript, allowing both administrators and users to navigate easily. Users can effortlessly browse movies, provide ratings, and access recommendations.
Efficient Movie Management for Admins: Admins have multiple functionalities, including adding movies, uploading and managing datasets, monitoring user ratings, and analyzing data trends through graphical insights. These features enhance content management and user engagement.
Graph-Based Analysis for Data Insights: The system includes dynamic graph visualizations that allow admins to monitor user rating trends by genre (e.g., Action, Comedy, Sci-Fi). This feature helps in understanding which genres are most preferred and how user preferences shift over time.
Enhanced User Engagement and Satisfaction: Since recommendations are personalized and continuously updated, users remain engaged with the system. The sentiment analysis feature further enhances decision-making, allowing users to explore movies based on both numerical ratings and textual reviews.
Robust Data Handling and Storage: The use of MySQL as the database ensures efficient storage, retrieval, and management of movie records, user ratings, and sentiment data. The system can handle a large dataset (4,000 movie records) while maintaining performance.
Scalability and Future Enhancements: The system is designed to be scalable, meaning additional features such as AI-based recommendations, improved sentiment analysis, or expanded movie datasets can be integrated in the future without significant architectural changes.
Improved Decision-Making for Users: Users are not only provided with movie recommendations but also get insights from sentiment analysis, helping them make better viewing choices. This combination of user ratings and critics’ opinions creates a more informed decision-making process.
Overall, the proposed system enhances the movie discovery experience, making it dynamic, personalized, and insightful while ensuring that user preferences are continually updated and recommendations remain relevant.

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

Online Movie Recommendation System