St.

MARTIN’S ENGINEERING COLLEGE

UGC Autonomous
NBA & NAAC A+ ACCREDITED
Dhulapally, Secunderabad– 500100

Department of Computer Science and Engineering

A new frame work for Movie Recommendation

Prediction using Supervised Learning Algorithm
Batch No: D17
1. K.Koushik Kumar (22K81A05M6)
2. P.Yashwanth Babu (22K81A05N4)
3. S.Ranjith (22K81A05P0)
4. J.Vijaya (22K81A0527)
Under the Guidance of
Dr. G. Jawaherlal Nehru
Associate Professor
Department of Computer Science and Engineering 1
 OUTLINE

1. Abstract
2. Introduction
3. Literature Survey
4. Existing System
5. Proposed System
6. Modules
7. References

2
 ABSTRACT
Users often face the problem of excessive available information. Recommendation systems (RSs)
are deployed to help users cope up with the information explosion. RS is mostly used in digital
entertainment, such as Netflix, Prime Video, and IMDB, and e-commerce portals such as Amazon,
Flipkart, and eBay.

In this article, we focus on RS for movies, which is an important source of recreation and
entertainment in our life. Movie suggestions for users depend on Web-based portals. Movies can
be easily differentiated through their genres, such as comedy, thriller, animation, and action.
Another possible way to categorize the movies based on its metadata, such as release year,
language, director, or cast. Most online video-streaming services , provide personalized user
experience by utilizing the user’s historical data, such as previously viewed or rated history.
3
 ABSTRACT

Internet Movie Database (IMDB) and Netflix database,5 that were not found suitable for
our work due to the absence of microblogging data. After a thorough assessment of the
abovementioned databases, the MovieTweetings database was finally selected for the
proposed system. MovieTweetings is widely considered as a modern version of the
MovieLens database. The purpose of this database is to provide an up-to-date movie rating
so that it contains more realistic data for sentiment analysis. Table I displays the relevant
details of the MovieTweetings database.

4
 INTRODUCTION

• Imagine scrolling through endless movie options on your favourite streaming platform, feeling
overwhelmed by the sheer number of choices. This is where recommendation systems come in
handy. They're like personal assistants, suggesting movies tailored just for you based on what
you've watched before.

• Today, we're diving into the world of movie recommendations. Movies are a big part of our
entertainment, but with so many out there, finding the perfect one can be like searching for a
needle in a haystack.

• In response to these challenges, this study introduces a novel framework for movie
recommendation prediction using supervised learning algorithms. 5
 INTRODUCTION

• By leveraging supervised learning techniques, including regression or classification algorithms,

the proposed framework aims to overcome the limitations of existing approaches and deliver
more accurate and personalized movie recommendations.

• Key Factors of this project is by analyzing how you've rated movies in the past and considering
factors like genre and release year, our system strives to give accurate suggestions.

• We primarily use Python for the front-end, HTML, CSS, and JavaScript for design, and MySQL
as the database

6
 LITERATURE SURVEY
S.
Author Title Year Contributions
No
1. Amazon.com Described item-to-item
Linden, G., Smith, B., recommendations: Item- collaborative filtering
& York, J. 2003
to-item collaborative for personalized
filtering recommendations.
2. Survey of neighborhood- Provided overview of
Desrosiers, C., &
based recommendation 2011 collaborative filtering
Karypis, G. methods and similarity measures.
3. Proposed neural
Modeling fashion trends network-based system
He, R., & McAuley, J. with collaborative filtering 2016
for capturing user-item
interactions.
7
 EXISTING SYSTEM
Various recommendation systems (RSs) have emerged in recent decades, employing different
methods like Collaborative Filtering (CF), Content-Based Filtering (CBF), and hybrid
approaches. CF, initially introduced for document search systems, inferred user preferences
from interactions, such as pages read. This helps mitigate the cold start problem. Optimizing
RSs poses challenges, leading to the proposal of algorithms like gray wolf optimization and
artificial bee colony. A collaborative movie RS was developed using gray wolf optimization
and fuzzy c-mean clustering, showing improved performance. Additionally, an artificial bee
colony and k-mean cluster framework were proposed to address scalability and cold start
issues in collaborative movie RSs.

8
 EXISTING SYSTEM

DISADVANTAGES OF EXISTING SYSTEM:

1. The existing users not only receive information according to their social links
but also gain access to other user-generated information.
2. The necessity of prior user history and habits for performing the task of
recommendation..

9
 PROPOSED SYSTEM
The system relies on two types of databases: one containing user-rated movies and another
containing user tweets from Twitter.

Public Databases:

1. Various public databases were explored, including Movielens 100K, Movielens 20M,
IMDb, and Netflix, but none contained suitable microblogging data.

2. After evaluation, the MovieTweetings database was chosen for its up-to-date movie
ratings, making it ideal for sentiment analysis.

Modified MovieTweetings Database:

1. The MovieTweetings database was adapted for RS implementation, focusing on utilizing

sentiment analysis of user tweets for movie recommendations.

2. Only movies released in or after 2014 were considered due to limited tweet availability
for older films. 10
 PROPOSED SYSTEM

ADVANTAGES OF PROPOSED SYSTEM:

1.To use movie tweets is to understand the current trends, public sentiment, and user
response of the movie.
2. Experiments conducted on the public database have yielded promising results.

11
 MODULES
Step-1:Data Collection and Preprocessing:
1. Gather movie ratings and metadata from diverse sources.
2. Clean and preprocess the dataset, handling missing values and encoding categorical
variables.

Step-2:Feature Engineering:
1. Extract relevant features from the dataset, including user characteristics and movie
attributes.
2. Perform feature transformation and normalization to prepare the data for
modelling.

Step-3:Model Selection and Training:

1. Choose appropriate supervised learning algorithms for movie recommendation
prediction.
2. Train the selected models using the preprocessed dataset and optimize
hyperparameters. 12
 MODULES

Step-4:Evaluation and Validation:

1. Evaluate the trained models using suitable evaluation metrics such as Mean
Absolute Error (MAE) or Root Mean Squared Error (RMSE).
2. Validate the performance of the models on a separate testing dataset to assess
generalization capabilities.

Step-5:Deployment and Integration:

1. Deploy the trained model into a production environment, integrating it with user
interfaces or streaming platforms.
2. Implement mechanisms for real-time recommendation generation and feedback
collection..
13
 REFERENCES
[1]Koren, Y., Bell, R., & Volinsky, C. (2009). Matrix factorization techniques for
recommender systems. IEEE Computer, 42(8), 30-37.
[2]Desrosiers, C., & Karypis, G. (2011). A comprehensive survey of neighborhood-
based recommendation methods. ACM Transactions on Internet Technology (TOIT),
11(4), 1-50.
[3]Adomavicius, G., & Tuzhilin, A. (2005). Toward the next generation of
recommender systems: A survey of the state-of-the-art and possible extensions. IEEE
Transactions on Knowledge and Data Engineering, 17(6), 734-749.
[4]He, R., & McAuley, J. (2016). Ups and downs: Modeling the visual evolution of
fashion trends with one-class collaborative filtering. Proceedings of the 25th
International Conference on World Wide Web (WWW '16), 507-517.
14
 REFERENCES
[5]Linden, G., Smith, B., & York, J. (2003). Amazon.com recommendations: Item-to-
item collaborative filtering. IEEE Internet Computing, 7(1), 76-80.
[6]Melville, P., Mooney, R. J., & Nagarajan, R. (2002). Content-boosted collaborative
filtering for improved recommendations. Proceedings of the Eighteenth National
Conference on Artificial Intelligence (AAAI-02), 187-192.
[7]Pazzani, M. J., & Billsus, D. (2007). Content-based recommendation systems. The
Adaptive Web, 3251, 325-341.
[8]Su, X., & Khoshgoftaar, T. M. (2009). A survey of collaborative filtering techniques.
Advances in Artificial Intelligence, 2009, 421425.
[9]Burke, R. (2002). Hybrid recommender systems: Survey and experiments. User
Modeling and User-Adapted Interaction, 12(4), 331-370. 15
QUERIES ??

16
THANK YOU

17