INTERNSHIP REPORT ON

“ARFICIAL INTELLIGENCE”
Project Report submitted to

BANGALORE UNIVERSITY

Bachelor of Computer Application By

ABHINAV K(U03CT21S0070)
UNDER THE GUIDANCE OF

Prof. ABHIJEET UPRETY

Department of Computer Science

EAST WEST COLLEGE OF MANAGEMENT
Off Magadi Road, Vishwaneedam Post, Bangalore 560091

YEAR 2023- 2024

1
 EAST WEST COLLEGE OF MANAGEMENT
Off Magadi Road, Vishwaneedam Post, Bangalore-560091

Department of Computer Applications

Certified that the internship work entitled “ARTIFICIAL INTELLIGENCE” is

a work carried out by ABHINAV K (U03CT21S0070) in partial fulfilment for the
award of Degree of Bachelor of Computer Application of the Bangalore
University, Bangalore during the year 2023- 2024.The internship report has been
approved as it satisfies the academic requirements in respect of Internship work
prescribed for the bachelor of Computer Applications.

Signature of the Guide Signature of the HOD Date:

Signature of the Examiner

1.

2.

2
 DECLARATION

We hereby declare that the work which is being presented in the internship
entitled “ARTIFICIAL INTELLIGANCE” in partial fulfilment of the
requirements for the award of the Degree of bachelor of Computer Application
submitted in the Department of Computer Science East West College of
Management is an authentic record of our own work carried out under the
supervision and guidance of “Prof. ABHIJEET UPRETY” in the department of
Computer Sciences, East West College of Management.

Date: ABHINAV K

Place: Bangalore U03CT21S0070

ACKNOWLEDGEMENT

I would like to express my deepest gratitude to EAST WEST COLLEGE OF MANAGEMENT

for providing me with the invaluable opportunity to undertake this internship. The resources,
support, and guidance offered by the college have been instrumental in my professional and
3
personal growth during this period. The commitment to excellence and the emphasis on real-
world experience at EAST WEST COLLEGE OF MANAGEMENT have equipped me with the
tools necessary to thrive in this internship and beyond.

My heartfelt thanks go to YHills Edutech Private Limited for granting me the chance to be a
part of their esteemed organization. The experience and knowledge I have gained during my
time at YHills Edutech have been beyond measure. I am profoundly grateful to my supervisor,
Anikesh, for his constant guidance, encouragement, and invaluable insights throughout the
internship. His patience, expertise, and mentorship have greatly contributed to my learning and
development. Anikesh's ability to provide constructive feedback and foster a positive learning
environment has been pivotal in my professional growth.

I am also deeply appreciative of Professor ABHIJEET UPRETY, whose continuous support and
encouragement have been a driving force in my academic journey. His advice and feedback have
been crucial in helping me navigate the challenges of this internship and in shaping my approach
to professional work. Professor Abhijeeth's dedication to student success and his willingness to
go above and beyond have been a source of inspiration for me.

Finally, I would like to thank my family and friends for their unwavering support and
encouragement. Their belief in my abilities and their emotional support have been pivotal in
helping me complete this internship successfully. The encouragement from my loved ones has
been a constant source of motivation, driving me to strive for excellence in all my endeavors.

Thank you all for your support, guidance, and encouragement. This internship has been a
transformative experience, and I am deeply grateful to everyone who has contributed to my
journey.

COMPANY CERTIFICATE

4
COMPANY PROFILE

5
Company Profile: YHills Edutech Private Limited

Overview:
YHills Edutech Private Limited is a pioneering educational technology company committed to
addressing the prevailing employment gap by equipping learners with essential entry-level skills
required by the job market. Through innovative digital solutions and personalized learning
experiences, YHills aims to empower individuals to make meaningful contributions to society
while honing their skills and perfecting their craft.

History:
YHills Edutech Private Limited was founded with a bold vision to revolutionize the way
individuals acquire skills and enter the workforce. Recognizing the disconnect between the skills
possessed by graduates and those demanded by employers, the company embarked on a mission
to bridge this gap and create opportunities for meaningful employment. Since its inception in
[year], YHills has been at the forefront of driving change in the education and employment
sectors, leveraging technology to empower learners and transform lives.

Mission:
YHills wants to change the status quo of prevailing employment scenario, where there is a wide
mis-match between graduates looking for jobs and open positions requiring certain set of entry-
level skills. Our mission is to provide learners with the practical skills and knowledge they need
to succeed in today's competitive job market, ultimately fostering economic empowerment and
social mobility.

Vision:
YHills's vision is deeply rooted in the tenet of "Teaching a Man How to Fish". We want learners
to provide meaningful contributions to society individually and collaboratively, honing and
perfecting their craft in the process. By equipping individuals with the skills and confidence they
need to thrive in their careers, we envision a future where every person has the opportunity to
achieve their full potential and make a positive impact on the world.

Products and Services:

YHills Edutech offers a comprehensive suite of products and services designed to meet the
diverse needs of learners at every stage of their educational journey. Our offerings include
interactive e-learning platforms that provide engaging and immersive learning experiences,
personalized skill development programs tailored to individual learning styles and goals,
industry-specific training modules that prepare learners for in-demand careers, career readiness
workshops that equip participants with essential soft skills and job search strategies, and
mentorship and coaching programs that provide guidance and support to help learners succeed.
6
Organizational Structure:
YHills Edutech operates with a dynamic and adaptive organizational structure, fostering a
culture of collaboration, innovation, and continuous learning. The company is structured into
specialized teams focused on product development, curriculum design, marketing, sales,
customer support, and research and development. Each team is led by experienced professionals
with expertise in their respective areas, ensuring that YHills remains at the forefront of
educational innovation and delivers high-quality products and services to our customers.

Culture and Values:

At YHills Edutech, our culture is guided by core values of innovation, integrity, and inclusivity.
We believe in fostering an environment where creativity thrives, diversity is celebrated, and
every individual is empowered to reach their full potential. Our commitment to excellence drives
us to constantly innovate and push the boundaries of what's possible in education. We are
dedicated to creating a supportive and inclusive work environment where all employees feel
valued, respected, and inspired to do their best work.

Impact and Achievements:

Since its inception, YHills Edutech has made significant strides in empowering learners and
bridging the skills gap in the job market. Our innovative approach to skill development has been
recognized for its effectiveness and impact, leading to meaningful employment opportunities for
countless individuals. We have partnered with leading educational institutions, businesses, and
organizations to deliver high-quality learning experiences to learners around the world. YHills
continues to make a positive difference in the lives of learners, empowering them to succeed in
their careers and contribute meaningfully to society.

Conclusion:

YHills Edutech Private Limited is more than just an educational technology company; it's a
catalyst for change in the way individuals acquire skills and enter the workforce. With a mission
to empower learners and a vision of creating a more equitable and inclusive job market, YHills is
7
committed to reshaping the future of education and employment. Through our innovative
products and services, dedicated team, and unwavering commitment to excellence, we are
transforming lives and empowering individuals to achieve their full potent

8
 ABSTRACT

This internship report encapsulates a transformative journey at YHills Edutech Private Limited,
spanning from April 4 , 2024 to May 5 , 2024, marking a pivotal chapter in my academic and
professional growth. Rooted in the ethos of bridging the gap between academic learning and
industry demands, this internship sought to immerse me in the dynamic realm of educational
technology while nurturing my skills and knowledge in alignment with my academic pursuits at
EAST WEST COLLEGE OF MANAGEMENT.

In recent years, the exponential growth of digital content has made it increasingly challenging
for users to discover movies that align with their preferences. To address this issue, this project
aims to develop a sophisticated movie recommendation system as part of a Bachelor of
Computer Applications (BCA) internship. This system leverages advanced data analysis
techniques and machine learning algorithms to provide personalized movie recommendations to
users.

The core objective of the recommendation system is to enhance user experience by predicting
and suggesting movies that match individual tastes and viewing habits. This is achieved through
the integration of collaborative filtering, content-based filtering, and hybrid approaches.
Collaborative filtering utilizes user behaviour and interaction data to find similarities between
users and suggest movies liked by similar users. Content-based filtering, on the other hand,
recommends movies based on the characteristics of the movies themselves, comparing them to a
user’s previously liked movies. The hybrid approach combines these methods to mitigate their
respective limitations and improve recommendation accuracy.

The development process involves several key stages, including data collection, preprocessing,
model training, and system evaluation. Publicly available movie datasets, such as the MovieLens
dataset, are employed for training and testing purposes. Various machine learning algorithms,
such as k-nearest neighbours, matrix factorization, and deep learning techniques, are explored
and compared to identify the most effective model.

The system is designed to be user-friendly, with an intuitive interface that allows users to easily
interact with the recommendation engine. Evaluation metrics such as precision, recall, and F1-
score are used to assess the performance and reliability of the recommendations.

This project not only aims to deliver a functional movie recommendation system but also
provides an insightful learning experience in machine learning, data analysis, and software
development. The outcomes of this internship project will contribute to the broader field of
recommendation systems and offer valuable insights for future research and development.

9
INDEX
10
 SL.N PG.NO
O
1 Learning objectives………………… 12-13
2 Weekly overview of internship activities 14-18
3 Introduction……………………………. 19
4 Modules DESCRIPTION……………… 20-21
5 Analyses and Test case…………………. 22-23
6 Hardware and software used………….. 24-25
7 Technology used……………………… 26
8 Screenshots…………………………… 27-28
9 Conclusion………………………………. 33

LEARNING OBJECTIVES
Technical Skills

1. Data Handling and Manipulation

11
• Learn to extract, clean, and preprocess data using tools like Pandas and NumPy.
• Gain proficiency in handling large datasets and performing data analysis.
2. Understanding Recommendation Algorithms
• Understand the theoretical foundations of recommendation algorithms, including collaborative
filtering, content-based filtering, and hybrid approaches.
• Learn to implement these algorithms using programming languages like Python or R.
3. Machine Learning Techniques
• Acquire knowledge of machine learning techniques relevant to recommendation systems, such
as matrix factorization, SVD, and neural networks.
• Develop skills in training, validating, and fine-tuning machine learning models.
4. Programming and Software Development
• Improve programming skills in languages such as Python, Java, or Scala.
• Learn to use software development tools and best practices, including version control (e.g.,
Git), code modularity, and testing.
5. Web Development
• Gain experience in front-end development using HTML, CSS, and JavaScript.
• Learn to use web development frameworks such as React, Angular, or Vue.js to create
interactive user interfaces.
6. Database Management
• Understand the basics of database management, including SQL and NoSQL databases.
• Learn to design and implement efficient database schemas for storing and querying
recommendation data.
7. System Deployment
• Learn to deploy applications on cloud platforms like AWS, Google Cloud, or Azure.
• Understand the principles of containerization (e.g., Docker) and orchestration (e.g.,
Kubernetes) for scalable deployment.

Analytical and Problem-Solving Skills

1. Algorithm Selection and Implementation

• Develop the ability to choose the appropriate recommendation algorithm based on the problem
context and data characteristics.
• Implement and compare different algorithms to determine the best-performing model.
2. Performance Evaluation
12
• Learn to evaluate recommendation systems using appropriate metrics such as precision, recall,
F1 score, and RMSE.
• Understand the trade-offs between different evaluation metrics and how they impact the user
experience.
3. Optimization Techniques
• Acquire skills in optimizing algorithms for performance, scalability, and accuracy.
• Learn to apply techniques such as hyperparameter tuning and cross-validation to improve
model performance.

Professional Development

1. Project Management
• Learn to plan, execute, and manage a technical project from start to finish.
• Gain experience in using project management tools like JIRA, Trello, or Asana.
2. Team Collaboration
• Develop the ability to work collaboratively within a team, participating in code reviews, team
meetings, and collaborative coding sessions.
• Learn to communicate effectively with team members and stakeholders.
3. Documentation
• Improve skills in documenting the development process, including writing technical
documentation and user guides.
• Learn to prepare clear and concise reports on project progress and outcomes.
4. Presentation Skills
• Develop the ability to present technical information to both technical and non-technical
audiences.
• Gain experience in creating and delivering presentations on project findings and results.

13
 WEEKLY OVERVIEW OF INTERNSHIP ACTIVITIES

Week Objecti Activitie

ves s

Week 1: •Understand the 1.Introduction and

Foundation and basics of Planning:
Data Preparation recommendation •Familiarize
systems. yourself with
•Gather and different types of
preprocess the recommendation
movie dataset. systems
(collaborative
filtering, content-
based, hybrid).
•Define the scope
and goals of the
project.
2.Dataset
Acquisition:
•Obtain a relevant
movie dataset
(e.g., MovieLens,
IMDb, or Netflix).
•Understand the
structure of the
dataset (users,
movies, ratings,
metadata).
3.Data Cleaning
and Exploration:
•Handle missing
values, duplicates,
and
inconsistencies.
•Perform
exploratory data
analysis (EDA) to
understand data
distribution and
identify patterns.
•Visualize data
using libraries like
14
 Matplotlib or
Seaborn.
4.Data
Preprocessing:
•Normalize and
preprocess data for
model input.
•Split the dataset
into training and
testing sets.

Week 2: Building •Develop basic 1.Collaborative

the recommendation Filtering:
Recommendation algorithms. •Implement user-
Models
•Evaluate and based and item-
compare their based
performance. collaborative
filtering.
•Use libraries such
as Surprise or
implicit.
2.Content-Based
Filtering:
•Create content-
based
recommendation
mod
•Use TF-IDF or
other vectorization
techniques for text
data.
3.Model
Evaluation:
•Evaluate models
using metrics like
RMSE, MAE,
precision, and
recall.
•Perform cross-
validation to
ensure model
robustness.
4.Hyperparameter
Tuning:
•Optimize model
performance

15
 through
hyperparameter
tuning.els using
movie
Week 3: •Explore 1.Matrix
Advanced advanced Factorization:
Techniques and recommendation •Implement matrix
Integration
techniques. factorization
•Integrate and techniques such as
refine models. Singular Value
Decomposition
(SVD).
Explore advanced
methods like ALS
(Alternating Least
Squares).
2.Hybrid Models:
•Combine
collaborative and
content-based
filtering to create
hybrid models.
•Experiment with
different hybrid
approaches (e.g.,
weighted,
switching, feature
combination).
3.Incorporating
Implicit Feedback:
•Use implicit
feedback (e.g.,
clicks, views) to
improve
recommendation
accuracy.
•Implement
implicit feedback
models and
evaluate their
performance.
4.Model
Integration:
•Integrate the best-
performing models
into a unified

16
 recommendation
system.
•Ensure seamless
switching or
combination of
different models.

Week 4: •Deploy the 1.System

Deployment and recommendation Deployment:
Finalization system. •Prepare the model
•Document and for deployment
present the (e.g., using Flask,
project. Django).
•Deploy the model
on a cloud
platform (e.g.,
AWS, Heroku).
2.User Interface:
•Develop a simple
user interface to
interact with the
recommendation
system.
•Ensure the UI is
user-friendly and
responsive.
3.Testing and
Evaluation:
•Conduct user
testing to gather
feedback on the
recommendation
system.
•Fine-tune the
system based on
feedback.
4.Documentation
and Presentation:
•Document the
entire project,
including
methodologies,
code, and results.
•Prepare a final
presentation to
showcase your

17
 work.
•Highlight key
findings,
challenges, and
future work.

INTRODUCTION

Lights, camera, action! Welcome to MOVIE RECCOMENDATION SYSTEM, where the magic
of cinema meets the power of personalization. In a world inundated with countless movie
options, finding the perfect film can be a daunting task. That's why we've created a revolutionary
platform that puts the spotlight on you, the viewer.

18
At MOVIE RECCOMENDATION SYSTEM, we understand that every movie buff is unique,
with their own set of tastes, preferences, and moods. Our mission is simple: to transform your
movie-watching experience into an unforgettable journey tailored just for you. With our
innovative recommendation engine and extensive database of films, we're here to guide you
through the vast cinematic landscape and help you discover hidden gems, timeless classics, and
everything in between.

But MOVIE RECCOMENDATION SYSTEM is more than just a movie website. It's a vibrant
community of fellow cinephiles, where you can connect, share, and explore your love of film
together. Whether you're a casual viewer or a hardcore film buff, there's something here for
everyone.

So, grab your popcorn, settle into your favourite spot on the couch, and let MOVIE
RECCOMENDATION SYSTEM be your ticket to a world of cinematic wonder. Welcome to the
future of movie discovery. Welcome to MOVIE RECCOMENDATION SYSTEM.

MODULE DESCRIPTION

1.User Authentication and Profile Management:

19
This module allows users to register, log in, and manage their profiles. It includes features such
as user registration, login, password management, and profile editing.

2. Movie Recommendation Engine:

This module is the heart of the website, utilizing advanced algorithms to suggest personalized
movie recommendations based on user preferences, viewing history, and ratings. It includes
features such as recommendation generation, algorithm customization, and real-time updates.

3. Search and Filtering:

This module enables users to search for movies based on various criteria such as genre, actor,
director, release year, and keywords. It includes features such as advanced search functionality,
filtering options, and sorting capabilities.

4. User Interaction and Engagement:

This module focuses on fostering user interaction and engagement within the community. It
includes features such as user reviews, ratings, comments, likes, and social sharing. Additionally,
it may include features like user recommendations and following other users.

5. Admin Panel:
This module provides administrators with tools to manage the website, including user accounts,
content moderation, and system settings. It includes features such as user management, content
moderation, analytics, and system configuration.

6. Movie Database Management:

This module is responsible for managing the movie database, including adding new movies,
updating existing entries, and maintaining data accuracy. It includes features such as movie
database management, data import/export, and data validation.

7. Responsive Design and Cross-Platform Compatibility:

This module ensures that the website is accessible and user-friendly across different devices and
screen sizes. It includes features such as responsive design implementation, mobile optimization,
and cross-browser compatibility testing.

8. Third-Party Integrations:

20
This module allows integration with external APIs and services to enhance the website's
functionality. It includes features such as integration with movie databases (e.g., IMDb, TMDb),
social media platforms, and analytics tools.

9. Security and Privacy:

This module focuses on ensuring the security and privacy of user data and transactions. It
includes features such as encryption, secure authentication, data validation, and compliance with
privacy regulations (e.g., GDPR, CCPA).

10. Feedback and Support:

This module provides users with channels to provide feedback, report issues, and seek
assistance. It includes features such as contact forms, help documentation, FAQs, and customer
support channels (e.g., email, chat, ticketing system).

ANALYSING AND TEST CASE

ANALYSIS

The analysis phase of the movie recommendation system project involves a detailed examination
of the data and the performance of various algorithms. This phase is crucial for understanding
the effectiveness of the recommendation techniques and identifying areas for improvement. The
analysis consists of the following steps:

1. Exploratory Data Analysis (EDA):

- Data Overview: Summarize the dataset, including the number of movies, users, and ratings.
Analyze the distribution of ratings, genres, and other relevant features.
21
- Data Visualization: Use visual tools such as histograms, bar charts, and scatter plots to
visualize the data distribution and identify patterns or anomalies.
- *Correlation Analysis: Investigate the relationships between different features, such as genres
and user ratings, to uncover hidden insights.

2. Algorithm Performance Analysis:

- Collaborative Filtering Performance: Evaluate user-based and item-based collaborative filtering
techniques by analyzing the similarity matrices and recommendation results.
- Content-Based Filtering Performance: Assess the effectiveness of content-based filtering by
analyzing feature extraction methods and similarity calculations.
- Hybrid Model Performance: Compare the performance of the hybrid model against individual
collaborative and content-based models to determine its effectiveness.

3. Model Evaluation Metrics:

- Precision: Measure the proportion of recommended movies that are relevant to the user.
- Recall: Measure the proportion of relevant movies that are successfully recommended to the
user.
- F1-Score: Calculate the harmonic mean of precision and recall to provide a single performance
metric.
- Mean Squared Error (MSE): Evaluate the accuracy of predicted ratings by calculating the
average squared difference between predicted and actual ratings.

4. Hyperparameter Tuning:
- Grid Search: Perform a grid search to find the optimal hyperparameters for each algorithm,
such as the number of neighbors in k-NN or the number of latent factors in matrix factorization.
- Cross-Validation: Use cross-validation techniques to validate the models and ensure they
generalize well to unseen data.

TEST CASE

1. User-Based Collaborative Filtering Test Cases:

- Test Case 1: Verify recommendations for a user with a high number of ratings.
- Input: User ID with a high number of movie ratings.
- Expected Output: A list of recommended movies based on similar users' preferences.
- Test Case 2: Verify recommendations for a new user with few ratings.
- Input: U
- Expected Output: A list of movies similar to the popular movie.
- Test Case 4: Verify recommendations for a less-known movie.
- Input: Movie ID of a less-known movie.
- Expected Output: A list of movies similar to the less-known movie.

22
2. Content-Based Filtering Test Cases:
- Test Case 5: Verify recommendations based on genre similarity.
- Input: User ID with high ratings for a specific genre.
- Expected Output: A list of movies from the same or similar genres.
- Test Case 6: Verify recommendations based on director similarity.
- Input: User ID with high ratings for movies by a specific director.
- Expected Output: A list of movies directed by the same or similar directors.

3. Hybrid Model Test Cases:

- Test Case 7: Verify hybrid recommendations for a user with diverse preferences.
- Input: User ID with diverse movie ratings across genres.
- Expected Output: A list of recommended movies that consider both user preferences and movie
content.
- Test Case 8: Verify hybrid recommendations for a user with specific preferences.
- Input: User ID with specific preferences for certain genres or directors.
- Expected Output: A list of recommended movies that align closely with the user's specific
preferences.

4. User Interface Test Cases:

- Test Case 9: Verify user registration and login functionality.
- Input: User registration details and login credentials.
- Expected Output: Successful user registration and login.
- Test Case 10: Verify movie rating and feedback functionality.
- Input: User ratings and feedback for movies.
- Expected Output: Successful submission and storage of ratings and feedback.

5. System Performance Test Cases:

- Test Case 11: Verify system response time for generating recommendations.
- Input: User ID requesting recommendations.
- Expected Output: Recommendations generated within an acceptable time frame.

HARDWARE AND SOFTWARE

Hardware Requirements

1. Development Environment:
• Processor: Intel Core i5 or equivalent (Quad-core) or higher.
• Memory (RAM): 8 GB or higher.
• Storage: 500 GB HDD or 256 GB SSD for faster read/write operations.
• Graphics Processing Unit (GPU): Optional but beneficial for training deep learning models
(e.g., NVIDIA GTX 1060 or higher).
23
2. Deployment Environment:
• Processor: Intel Core i5 or equivalent (Quad-core) or higher.
• Memory (RAM): 16 GB or higher for handling multiple user requests.
• Storage: 1 TB HDD or 512 GB SSD for efficient data storage and retrieval.
• GPU: Optional for real-time model inference (e.g., NVIDIA Tesla T4).

Software Requirements

• Frameworks: Flask/Django (backend), React/Angular (frontend)

• Version Cont
1. Operating System:
• Development: Windows 10/11, macOS, or Linux (Ubuntu 20.04 LTS or higher).
• Deployment: Linux (Ubuntu 20.04 LTS or higher) preferred for server environment.
2. Programming Languages:
• Python: Primary language for data processing, model training, and backend development.
• JavaScript: For frontend development and user interface interactions.
3. Development Tools and Libraries:
• Integrated Development Environment (IDE):
• PyCharm: For Python development.
• Visual Studio Code: For both Python and JavaScript development.
• Jupyter Notebook: For data exploration and prototyping machine learning models.
4. Data Science and Machine Learning Libraries:
• NumPy: For numerical computations.
• Pandas: For data manipulation and analysis.
• Scikit-learn: For implementing machine learning algorithms.
• SciPy: For scientific computing.
• TensorFlow or PyTorch: For deep learning models.
• Surprise: For building and evaluating recommender systems.
5. Database:
• SQLite: For lightweight, file-based database during development.
• MySQL or PostgreSQL: For more robust database management in deployment environments.
6. Web Development Frameworks:
• Flask or Django: For backend development and API creation.
• React or Angular: For frontend development to create a dynamic and responsive user interface.
7. Version Control:
• Git: For version control and collaboration.
• GitHub or GitLab: For repository hosting and project management.
8. Deployment and Hosting:
• Docker: For containerization to ensure consistent deployment across different environments.
• NGINX: For web server and reverse proxy.
• AWS, Google Cloud Platform (GCP), or Microsoft Azure: For cloud deployment and
scalability.
• Heroku: For simple deployment of smaller projects.
9. Miscellaneous Tools:

24
• Anaconda: For managing Python packages and virtual environments.
• Postman: For testing APIs.
• Jenkins or GitHub Actions: For continuous integration and continuous deployment (CI/CD).

Summary:

Development Hardware:

•Intel Core i5, 8 GB RAM, 500 GB HDD/256 GB SSD.

Deployment Hardware:

•Intel Core i5, 16 GB RAM, 1 TB HDD/512 GB SSD.

Software:

•OS: Windows 10/11, macOS, or Linux.

•Languages: Python, JavaScript.
•IDE: PyCharm, Visual Studio Code.
•Libraries: NumPy, Pandas, Scikit-learn, TensorFlow

TECHNOLOGY USED

•Languages: Python, JavaScript

•Libraries: NumPy, Pandas, Scikit-learn, SciPy, TensorFlow/PyTorch, Surprise
•IDE/Tools: PyCharm, Visual Studio Code, Jupyter Notebook
•Databases: SQLite (development), MySQL/PostgreSQL (production)
rol: Git, GitHub/GitLab
•Containerization: Docker
25
•Deployment: NGINX, AWS/GCP/Azure, Heroku
•Miscellaneous: Anaconda, Postman, Jenkins/GitHub Actions

SCREENSHOTS

Screenshot1

26
Screenshot2

27
PROJECT CODE

28
pip install pandas numpy scikit-learn flask

import pandas as pd

# Load datasets
movies = pd.read_csv('movies.csv')
ratings = pd.read_csv('ratings.csv')

# Merge datasets
data = pd.merge(ratings, movies, on='movieId')

# Display the first few rows of the dataset

print(data.head())

import matplotlib.pyplot as plt

import seaborn as sns

# Number of ratings per movie

movie_ratings = data.groupby('title')['rating'].count().sort_values(ascending=False)

# Plot the number of ratings per movie

plt.figure(figsize=(10, 6))
sns.histplot(movie_ratings, bins=50, kde=False)
plt.xlabel('Number of Ratings')
plt.ylabel('Count')
plt.title('Number of Ratings per Movie')
plt.show()

# Average rating per movie

average_ratings = data.groupby('title')['rating'].mean().sort_values(ascending=False)

29
# Plot the average ratings per movie
plt.figure(figsize=(10, 6))
sns.histplot(average_ratings, bins=50, kde=False)
plt.xlabel('Average Rating')
plt.ylabel('Count')
plt.title('Average Ratings per Movie')
plt.show()

from sklearn.metrics.pairwise import cosine_similarity

# Create a pivot table with users as rows, movies as columns, and ratings as values
user_movie_matrix = data.pivot_table(index='userId', columns='title', values='rating')

# Fill NaN values with 0

user_movie_matrix.fillna(0, inplace=True)

# Compute the cosine similarity between users

user_similarity = cosine_similarity(user_movie_matrix)

# Create a DataFrame for user similarity

user_similarity_df = pd.DataFrame(user_similarity, index=user_movie_matrix.index,
columns=user_movie_matrix.index)

def get_user_based_recommendations(user_id, num_recommendations=5):

# Get the similarity scores for the given user
user_sim_scores = user_similarity_df[user_id].sort_values(ascending=False)

# Get the indices of the users with the highest similarity scores
similar_users = user_sim_scores.index[1:num_recommendations+1]

30
 # Get the movies rated by the similar users
similar_users_movies = user_movie_matrix.loc[similar_users]

# Compute the average ratings for the movies rated by similar users
avg_ratings = similar_users_movies.mean(axis=0)

# Get the movies not yet rated by the target user

user_movies = user_movie_matrix.loc[user_id]
movies_not_rated = user_movies[user_movies == 0]

# Filter the average ratings to include only movies not rated by the target user
recommendations = avg_ratings[movies_not_rated.index].sort_values(ascending=False)

return recommendations.head(num_recommendations)

# Get recommendations for a specific user

print(get_user_based_recommendations(1))

from sklearn.feature_extraction.text import TfidfVectorizer

# Create a TF-IDF matrix for the movie genres

tfidf = TfidfVectorizer(stop_words='english')
tfidf_matrix = tfidf.fit_transform(movies['genres'])

# Compute the cosine similarity between movies

movie_similarity = cosine_similarity(tfidf_matrix, tfidf_matrix)

# Create a DataFrame for movie similarity

movie_similarity_df = pd.DataFrame(movie_similarity, index=movies['title'],
columns=movies['title'])

def get_content_based_recommendations(movie_title, num_recommendations=5):

31
 # Get the similarity scores for the given movie
movie_sim_scores = movie_similarity_df[movie_title].sort_values(ascending=False)

# Get the indices of the movies with the highest similarity scores
recommended_movies = movie_sim_scores.index[1:num_recommendations+1]

return recommended_movies

# Get recommendations for a specific movie

print(get_content_based_recommendations('Toy Story (1995)'))

def get_hybrid_recommendations(user_id, movie_title, num_recommendations=5):

user_recommendations = get_user_based_recommendations(user_id, num_recommendations)
content_recommendations = get_content_based_recommendations(movie_title,
num_recommendations)

# Combine the recommendations

hybrid_recommendations = pd.concat([user_recommendations,
content_recommendations]).drop_duplicates().head(num_recommendations)

return hybrid_recommendations

CONCLUSION

32
In closing, movie recommendation system emerges as a beacon of innovation and convenience
in the realm of movie discovery. Our platform seamlessly merges cutting-edge technology with
user-centric design, offering a personalized journey through the vast world of cinema. With a
robust recommendation engine at its core, coupled with intuitive search functionalities and user
engagement features, movie recommendation systems empower movie enthusiasts to explore,
discover, and share their favourite films with ease.

Throughout the development process, our focus remained steadfast on delivering an exceptional
user experience across all devices. From desktops to smartphones, movie recommendation
system ensures accessibility and usability without compromise, allowing users to indulge in their
cinematic passions anytime, anywhere. Moreover, our commitment to fostering a vibrant
community of film lovers sets us apart, providing a space for meaningful interactions,
discussions, and shared experiences.

Looking ahead, movie recommendation system stands poised to continue its evolution, driven by
a dedication to excellence and a passion for enhancing the movie-watching experience. As we
embark on this journey into the future, we extend our gratitude to our dedicated team and loyal
users who have joined us in redefining the way we engage with cinema. Together, we've built
more than just a website—we've created a home for cinephiles, a sanctuary for movie lovers, and
a destination for unforgettable cinematic adventures. Welcome to the next chapter of movie
discovery. Welcome to movie recommendation system.

33