BHARTI VIDYAPEETH INSTITUTE OF TECHNOLOGY

Sector-7, C.B.D. Belapur, Navi Mumbai

Certificate
This is to certify that the report of project entitled,

“Movie Recommendation System Using ML”

Submitted by:
2100270146 3525 Samarth Arote
2100270051 3526 Varsha Raut
2100270177 3527 Krupali Wayal
2200270646 3528 Aryan Patil
To The
MAHARASHTRA STATE BOARD OF TECHNICAL EDUCATION
In partial fulfillment of the requirement of the Course of

DIPLOMA IN INFORMATION TECHNOLOGY

Has been satisfactorily carried out under the guidance of

PROF MR. RANJEET PAWAR

In SEM V during the academic year 2023

~1~
 ACKNOWLEDGEMENT
We would like to express our heartfelt gratitude and appreciation to
Prof. P.N. Tandon, Principal of Bharati Vidyapeeth Institute of
Technology, Kharghar, for their unwavering support and guidance
throughout the development of our Movie Recommendation System
Using Machine Learning.
We are also deeply thankful to Prof. Ranjeet Pawar, Head of the
Department, Information Technology, for their invaluable mentorship
and continuous support.
Their commitment to promoting academic excellence and their
dedication to the growth of students have been a constant source of
inspiration for our team. We are honored to have had the opportunity to
work under their guidance and benefit from their vast knowledge and
experience.
This project would not have been possible without their unwavering
support, and for that, we extend our sincere gratitude.
Thank you for your belief in us and for being a part of our academic
journey.
Sincerely,
1. Samarth Arote 3525
2. Varsha Raut 3526
3. Krupali Wayal 3527
4. Aryan Patil 3528

~2~
 RATIONALE
Implementing a Movie Recommendation System Using Machine
Learning (ML) is a strategic decision driven by several compelling
rationales. Firstly, the film industry is characterized by a vast and diverse
collection of movies, making it challenging for users to discover content
aligned with their tastes. Machine learning algorithms can analyze user
behavior and preferences, providing personalized recommendations that
enhance the user experience. This leads to increased user engagement,
retention, and overall satisfaction. The recommendation system
leverages ML's ability to process and understand complex patterns within
vast datasets, enabling the delivery of tailored movie suggestions.
Secondly, the Movie Recommendation System addresses the need for
content personalization in the digital streaming era. With the rise of
online platforms, users expect tailored experiences. Machine learning
allows the system to consider various factors, such as user ratings,
viewing history, genre preferences, and social interactions, to deliver
recommendations that align with individual tastes. By continually
learning from user interactions, ML-based recommendation systems can
adapt to changing preferences, making them valuable tools in retaining
users and driving platform loyalty.

~3~
 ABSTRACT
Everyone loves movies no matter age, gender, race, colour, or
geographical location. We tend to all in the simplest way are connected
to every different via this wonderful medium, nonetheless what most
attention-grabbing is that the undeniable fact that however distinctive
our selections and combos are in terms of picture show preference. Some
individuals like genre-specific movies be it a thriller, romance, or sci-fi.
Whereas others specialize in lead actors and administrators. After we
take all the under consideration, it’s astoundingly troublesome to
generalize a movie and say that everybody would love it.
However, with all that said, it’s still seen that similar movies are liked
by a selected part of the society. So here’s whether we tend to as
information scientists get play and extract the juice out of all the
behavioural patterns of not solely the audience however conjointly from
the films themselves. Thus, while not additional ruction let’s jump right
into the fundamentals of a recommendation system. This paper is
planned a machine learning approach to suggest movies to the users
using KMeans clustering algorithm, K Nearest neighbours algorithm and
Affinity propagation clustering algorithm to recommend movies to the
users.

~4~
 INDEX

SR. NO. CONTENTS PAGE NO.

1. INTRODUCTION 6
2. LITERATURE SURVEY 7
3. SCOPE OF THE PROJECT 9
4. PROBLEM 12
DEFINITION/STATEMENT
5. PROPOSED METHODOLOGY 13
6. REQUIREMENT- 16
SOFTWARE/HARDWARE
7. DESIGN 16
8. IMPLEMENTATION 19
9. RESULT 21
10. TESTING 22
11. APPLICATION 24
12. CONCLUSION 27
13. FUTURE SCOPE 27
14. REFERENCES 29

INTRODUCTION

~5~
 Machine Learning is that field of study that offers computers the
aptitude to find out while not being expressly programmed. ML is one of
the foremost exciting technologies that one would have ever stumbled
upon because it is obvious from the name, it provides the pc that creates
it additional like humans: the power to learn. Machine Learning is actively
getting used these days, maybe in many places than one would expect.
Machine Learning is employed in net search engines, email filters to
delineated spam, websites to create individualized recommendations,
banking software systems to sight uncommon transactions, and much of
apps on our phones like voice recognition. Recommender systems are
systems that are designed to suggest things to the user that support
many alternative factors. These systems predict the foremost possible
product that the users are possibly to buy and are of interest to firms like
Netflix, Amazon etc. use recommender systems to assist their users to
spot the right product or movies for them. The recommender system
deals with an outsized volume data present by filtering the foremost
necessary information supported by the information provided by a user’s
preference and interest. It finds out the match between user and item
and imputes the similarities between users and ratings for
recommendation. Both the users and the services provided have
benefited from these sorts of systems, the standard and decision-making
method has additionally improved through these sorts of systems. In our
project, by exploring different Machine learning algorithm such as K-
Means clustering algorithm.

LITERATURE SURVEY
~6~
SR. TITLE YEA AIM DATASET METHOD RESULT
NO. R

1. Movie 2018 This paper MovieLens K Means System showed

recommendation proposed a (publicly clustering 95% accuracy on
System using machine Available) average in
clustering learning 11 target predicting rating
Algorithm and approach to class And from new user
Pattern recommend 12000 Neural data which can
recognition movies to users Network be used to 146
network users using K- analyze which
means movie should be
clustering recommended
algorithm to to new users.
separate
similar users
and creating
a neural
network for
each cluster.
2. TV series 2017 Predict what Movie data K Means 1st TV series
recommendation rating a user collected Clustering recommendation
Using fuzzy might give to from system that
reference certain TV by MovieLens consider no. of
system, KMean analyzing TV series Adaptive TV series as an
clustering and Imformation from IMDB Fuzzy input.
Adaptive neuro about user neuro
fuzzy inference and TV series. inference
system. system
(AFNIS).

~7~
3. Analysis of Movie 2020 To shows Movie- collaborati Movie
Recommendation that low Lens100k. ve filteringpredictions for
Systems; with rated movies the user with
and without are not user-Id 254: it is
considering the significant in Pearson observed that
low rated movies finding the correlation there is no
movie coefficient significant
predictions. difference
So it’s between the
suggestable predictions. The
to ignore negligible
them while difference
calculating between the
movie predictions
predictions shows that the
effect of
removing
lowrated movies
is negligible and
hence can be
removed.

4. Group 2008 To improve The hierarchica While doing

Recommendation quality of dataset l clustering clustering
System for service for used in this and accuracy
Facebook Facebook research decision improved by 9%
users, we was tree.
developed collected
GRS to find using
the most Facebook
suitable Platform.
group to join
by matching
users profiles
with groups
identity.

~8~
 SCOPE OF THE PROJECT
1. Data Collection and Preprocessing:
Gather movie-related data from various sources, including movie
databases, user ratings, reviews, and demographic information.
Preprocess the data to handle missing values, outliers, and ensure data
quality.
2. User Profiling:
Create user profiles by analyzing user behavior and preferences. This may
involve user ratings, movie genres, viewing history, and demographic
data.
3. Movie Profiling:
Develop movie profiles by analyzing movie attributes such as genre,
director, actors, release year, and user ratings.
4. Recommendation Algorithms:
Implement and experiment with various recommendation algorithms,
including collaborative filtering (user-based, item-based), content-based
filtering, and hybrid methods.
Explore deep learning models like neural collaborative filtering (NCF) or
matrix factorization for improved accuracy.
5. Real-time and Batch Processing:
Implement real-time recommendation systems to provide on-the-fly
suggestions to users during their browsing sessions.Develop batch

~9~
processing mechanisms to update recommendations periodically based
on evolving user preferences.
6. Evaluation Metrics:
Utilize relevant evaluation metrics such as precision, recall, F1-score,
RMSE (Root Mean Square Error), and NDCG (Normalized Discounted
Cumulative Gain) to assess the effectiveness of the recommendation
system.
7. User Interface:
Design a user-friendly interface, which may include a web or mobile
application, where users can access movie recommendations.
8. Personalization:
Focus on personalization by offering tailored recommendations to
individual users based on their historical data and behavior.
9. Scalability and Performance:
Ensure the recommendation system is scalable to handle a large user
base and extensive movie libraries without compromising performance.
10. Integration with External Data Sources:
- Integrate external data sources, such as social media, to incorporate
user interactions and feedback into the recommendation process.

~ 10 ~
11. Feedback Loop:
- Implement mechanisms for gathering user feedback on
recommendations and using this feedback to improve the system's
accuracy.
12. Privacy and Security:
- Address privacy concerns by ensuring that user data is handled securely
and in compliance with data protection regulations.
13. Maintenance and Updates:
- Plan for regular maintenance and updates, including retraining the ML
models and keeping movie databases up to date.
14. Testing and Validation:
- Conduct thorough testing and validation of the recommendation system
to ensure it meets the desired accuracy and performance standards.
15. Documentation:
- Create comprehensive documentation for developers, users, and
administrators to understand and use the system effectively.
16. User Feedback Analysis:
- Analyze user feedback and engagement to fine-tune the
recommendation algorithms and user interface.
17. Business Model and Monetization (Optional):
- Consider implementing a business model that may involve subscription
services, ad-based revenue, or partnerships with streaming platforms.
~ 11 ~
18. Future Enhancements:
- Identify opportunities for future enhancements, such as incorporating
natural language processing for content analysis, exploring reinforcement
learning, or enhancing the recommendation engine with additional data
sources.

PROBLEM DEFINITION/STATEMENT
The goal is to develop a movie recommendation system using machine
learning that effectively addresses the following challenges:
Information Overload: With an immense collection of movies across
various genres and languages, users are often overwhelmed and struggle
to make informed choices. The recommendation system must assist users
in navigating this vast content library by suggesting movies that align with
their interests.
Personalization: Every user has unique tastes and preferences. The
recommendation system should be capable of personalizing movie
suggestions based on a user's viewing history, ratings, and genre
preferences, providing a tailored experience for each user.
Cold Start Problem: For new users with limited or no viewing history, the
system must overcome the "cold start" problem by providing relevant
recommendations based on user profiles, demographics, or other
indicators.

~ 12 ~
 PROPOSED METHODOLOGY
1. Data Collection:
Gather movie-related data from various sources. This can include user
ratings, movie descriptions, genre information, and user profiles. You can
use datasets like MovieLens, IMDB, or custom data collection.
2. Data Preprocessing:
Clean and preprocess the data. This may involve handling missing values,
removing duplicates, and formatting the data for compatibility with the
machine learning models.
3. Feature Engineering:
Extract relevant features from the dataset. This can include user behavior
(e.g., ratings, watch history) and movie attributes (e.g., genre, director,
release year).
4. Data Splitting:
Divide the dataset into training, validation, and test sets. The training set
is used to build the model, the validation set helps in hyperparameter
tuning, and the test set is used to evaluate the final model's
performance.
5. Machine Learning Models:
Choose and implement machine learning models suitable for
recommendation systems. Common algorithms include:
Collaborative Filtering (user-based, item-based, matrix factorization)

~ 13 ~
Content-Based Filtering
Hybrid Models (combining collaborative and content-based approaches)
Deep Learning (e.g., Neural Collaborative Filtering)
6. Model Training:
Train the selected machine learning models on the training data.
Optimize hyperparameters and use techniques like cross-validation to
ensure model accuracy.
7. Model Evaluation:
Evaluate model performance using the validation set. Common
evaluation metrics include Mean Absolute Error (MAE), Root Mean
Square Error (RMSE), Precision, Recall, and F1-score.
8. Hyperparameter Tuning:
Fine-tune model hyperparameters to improve performance. Grid search
or random search can be employed for this purpose.
9. Model Selection:
Choose the best-performing model based on evaluation results from the
validation set.
10. Testing:
- Assess the final model on the test dataset to ensure its performance in
real-world scenarios.

~ 14 ~
11. User Interface Development:
- Create a user-friendly interface for users to interact with the
recommendation system. This can be a website, app, or command-line
interface.
12. Integration:
- Integrate the machine learning model with the user interface, ensuring
it provides recommendations to users based on their preferences.
13. Deployment:
- Deploy the recommendation system in a production environment,
making it accessible to users.
14. Feedback and Continuous Improvement:
- Collect user feedback to further enhance the system. Monitor its
performance and make improvements as needed.
15. Documentation:
- Prepare comprehensive documentation covering the methodology, data
sources, models used, and how to interact with the system.
16. Maintenance and Updates:
- Regularly update and maintain the recommendation system, adding
new features, and keeping it up-to-date with the latest data.

~ 15 ~
 REQUIREMENT-SOFTWARE/HARDWARE
Hardware Requirements:
 System: Recommended Intel core i3 3.3GHz and more.
 RAM: Minimum 4 GB, recommended 4GB or more.
 Hard disk: Minimum 200GB, recommended 500GB or more.
Software Requiremnts:
 Operating System (OS): Windows 7 and Above.
 Language used: Python .
 Dataset: CSV file
 Libraries: Pandas, Numpy, seaborn , matplotlib , TfidVectorizer,
cosine_similarity.
 Google Colab (or) Jupyter Notebook

DESIGN
DFD LEVEL 0:

~ 16 ~
DFD Level 1:

SEQUENCE DIAGRAM:

~ 17 ~
FLOWCHART:

~ 18 ~
 IMPLEMENTATION
In this segment, we have explained the way to be implemented in our
movie recommendation engine / system using python
programming language using K-Means clustering [31] library and K-
Nearest Neighbor. The implementation of the system
consists of many sub-sections which are standard processes to be
followed while solving any machine learning [32, 33, 34] problem.
These are as follows:
A. Dataset
B. Data Cleaning
C. Model Analysis
D. Model Building
A. Dataset
The movie dataset is hired in our research paper and collected from the
Kaggle database. The Kaggle [35, 36] database provides
datasets in the form of several varieties of movie content. The user rating
data consists of 211231 records and has a user ID,
movie ID, rating, and timestamp. The Characterization of the movie's
content information includes over 54058 records and
includes movie ID, title, genre, director, actor, and more.

~ 19 ~
B. Data Cleaning
Description of movies content data consists approximately 40 columns.
Most of those columns were not necessary for our
research paper and were therefore removed. The dataset contains plenty
of empty values and duplicate values that need to be
resolved. Additionally, there have been some entries for movies within
the user rating files from movies, which did not
correspond to any movies within description of movie content data.
These entries were erased for easy processing.
C. Data Analysis
We evaluated datasets to gain insight into movie datasets from Kaggle
database that can help to develop and perform our
engine / system using Matplotlib libraries in python programming. We
acknowledged patterns for most rated movies, most
rated genres, the number of movies in each genre, and the number of
rated movies in each rating category as shown below.

~ 20 ~
D. Model Building
We used the Python library to create a recommendation system . For
user-based filtering and item-based filtering, we used the user similarity
class for Pearson correlation similarity, which uses Pearson correlation to
work out similarity between users' ratings; Hence priority.

RESULT
The success of a movie recommendation system can be measured by
several key outcomes:
Improved User Engagement: Users spend more time on the platform,
watch more movies, and interact more with recommended content.
Increased Click-Through Rate (CTR): Users are more likely to click on
recommended movies, indicating that the recommendations are relevant
and appealing.
Higher User Ratings: Users rate recommended movies positively,
resulting in improved user satisfaction and loyalty.
Diverse Suggestions: The system provides diverse recommendations that
align with different user preferences and tastes.

~ 21 ~
Reduced Churn: The recommendation system helps retain users by
keeping them engaged with the platform.

TESTING
Test Data Preparation:
Prepare a diverse dataset of user preferences, movie ratings, and movie
features. This dataset should be representative of the target user base.
Offline Testing:
Conduct offline testing to assess the recommendation system's
performance using historical data. Common evaluation metrics include
Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and
Precision at K.
A/B Testing:
Deploy the recommendation system in a controlled environment where
some users receive recommendations from the new system, while others
receive recommendations from the old system
Online Testing:
Monitor the system's performance in real-time. Evaluate key
performance indicators such as user engagement, click-through rate, and
conversion rate for recommended movies.
User Feedback and Surveys:
Collect feedback from users regarding the quality of recommendations.
Use surveys and user reviews to gather qualitative insights.

~ 22 ~
Cold Start Testing:
Assess the system's ability to provide recommendations to new users
with little to no historical data. Measure the system's performance during
the "cold start" phase.
Exploration vs. Exploitation Testing:
Evaluate the balance between exploring new content and exploiting
known user preferences. Test different exploration strategies.
Dynamic Adaptation Testing:
Analyze how well the system adapts to changing user preferences and
external factors, such as trending movies or seasonal content.
Diversity Testing:
Assess the diversity of recommendations to prevent filter bubbles and
encourage users to explore a wider range of movies.
Robustness and Scalability Testing:
Ensure that the system can handle increased user activity and data
volume without a significant drop in performance.
Privacy and Security Testing:
Test the system's ability to protect user data and privacy, especially if it
involves sensitive health or personal information.
Bias and Fairness Testing:
Analyze the system for biases in recommendations, ensuring that it
provides fair and equitable suggestions for all user demographics.
~ 23 ~
Quality of Service (QoS) Testing:
Monitor the response times, availability, and system stability, especially
during peak usage periods.
Cross-Validation:
Use techniques like k-fold cross-validation to assess the model's
generalization performance on different subsets of the dataset.
Regression Testing:
Continuously test the system as it evolves and undergoes updates to
maintain or improve recommendation quality.
Benchmark Testing:
Compare the system's performance against industry benchmarks and
competitors to ensure it remains competitive.
Documentation and Reporting:
Document the testing procedures, results, and any issues encountered.
Create reports that highlight strengths, weaknesses, and potential areas
for improvement.

APPLICATION
Entertainment Platforms (Netflix, Amazon Prime, etc..):
Movie recommendation systems are widely used on streaming
platforms to suggest personalized movies and TV shows to users. ML
algorithms analyze user behavior, watch history, and preferences to

~ 24 ~
provide tailored recommendations. This enhances user engagement and
helps users discover content they might enjoy.
E-Commerce Platforms:
E-commerce websites like Amazon often use ML-based recommendation
systems to suggest movies for purchase or rental based on a user's
browsing and purchase history. This extends user engagement and
increases sales.
Social Media:
Social media platforms can employ movie recommendation systems to
recommend movies or share users' recent viewing activity. Friends can
see what others are watching and get recommendations based on their
network's choices.
Offline Movie Stores:
Physical movie rental stores or DVD rental kiosks can benefit from ML
recommendations by suggesting movies to customers based on their
rental history, preferences, and trends.
Movie Review Websites:
Websites like IMDb can provide movie recommendations to users
based on their ratings and reviews of other films. ML can help match
users with movies that align with their tastes.
Video-On-Demand Services:
In addition to streaming platforms, video-on-demand services can employ
movie recommendation systems to enhance user experiences,
~ 25 ~
encourage content discovery, and reduce churn rates.
Cinema Chain Websites:
Websites for cinema chains can use recommendation systems to
suggest movies for upcoming shows and help users choose the most
suitable films based on their viewing history.
Content Creation:
ML-driven movie recommendation can assist content creators and
filmmakers in identifying popular genres and themes, helping them make
informed decisions about what kind of movies to produce.
Advertising:
Advertisers can use recommendation systems to display movie-related
ads to users based on their preferences and behavior, leading to more
targeted advertising and higher conversion rates.
Subscription Services:
Subscription-based movie clubs or services can leverage ML
recommendations to curate personalized movie collections for
subscribers and improve retention rates.
Offline Events:
Movie recommendation systems can help organize offline events like
film festivals by suggesting films based on audience interests and reviews.

CONCLUSION
~ 26 ~
In this project, we worked on a K-Means clustering algorithm , knn
algorithm and Affinity Propagation Clustering Algorithm to make movie
recommendations as good as possible. User rating and preference have
been considered while building the system. Our system will calculate the
common in predicting rating from user information which may be used to
analyze that movie ought to suggest to new users using three Machine
Learning Algorithm. This proves that our system may be a valid one for
predicting within the field of movies. Finally, we compare the result of
the three algorithm and found that we got nearly same result with
different execution time. So we conclude that Affinity Propagation
Clustering Algorithm is more efficient than the other two algorithm.

FUTURE SCOPE
Neural Networks and Deep Learning have been all the rage the last
couple of years in many different fields, and it appears that they are also
helpful for solving recommendation system problems. Ben Allison, a
Principal Machine Learning Scientist at Amazon, gave a great talk earlier
this year at Amazon’s re:MARS conference about building recommender
systems using Recurrent Neural Networks and Deep Learning.
One of the benefits of Deep Learning is similar to matrix factorization,
in that there is an ability to derive latent attributes. Deep Learning,
however, can make up for some of the weaknesses of matrix
factorization such as the inability to include time in the model — which
standard matrix factorization isn’t designed for. Deep Learning, however,
can utilize Recurrent Neural Networks which are specifically designed for
time and sequence data.

~ 27 ~
 Incorporating time into a recommender system is important, because
there are often preference seasonal effects. For example, it is likely that
in December, more people are going to be watching holidaythemed
movies and buying home decorations. Another point that Ben Allison
brought up is the need to see what would happen if a customer was
shown a sub-optimal recommendation. This is taking a reinforcement
learning approach, since the goal in this case would be to show
customers a recommendation, and then record what the customer does.
At times, customers can be recommended something that does not
seem like the best option, just to see how the customer reacts which will
improve the learning in the long-term. Recommender systems can be a
very powerful tool in a company’s arsenal, and future developments are
going to increase business value even further. Some of the applications
include being able to anticipate seasonal purchases based on
recommendations, determine important purchases, and give better
recommendations to customers which can increase retention and brand
loyalty. Most businesses will have some use for recommender systems,
and I encourage everyone to learn more about this fascinating area.

~ 28 ~
 REFERENCES
[1] Chen, Hung-Chen, and Arbee L. P. Chen. “A music recommendation
system based on music data grouping and user interests.” Proceedings of
the tenth international conference on Information and knowledge
management - CIKM01, 2001, doi:10.1145/502585.502625.
[2] Ahmed, Muyeed, et al. “TV Series Recommendation Using Fuzzy
Inference System, K-Means Clustering and Adaptive Neuro Fuzzy
Inference System.” 2017, pp. 1512–1519.
[3] Park, Moon-Hee, et al. “Location-Based Recommendation System
Using Bayesian User‟s Preference Model in Mobile Devices.” Ubiquitous
Intelligence and Computing Lecture Notes in Computer Science, pp.
1130–1139., doi:10.1007/978-3-540-73549-6_110.
[4] Huang, Yao-Chang, and Shyh-Kang Jenor. “An audio recommendation
system based on audio signature description scheme in MPEG-7 Audio.”
2004 IEEE International Conference on Multimedia and Expo (ICME) (IEEE
Cat. No.04TH8763), doi:10.1109/icme.2004.1394273.
[5] Baatarjav, Enkh-Amgalan, et al. “Group Recommendation System for
Facebook.” On the Move to Meaningful Internet Systems: OTM 2008
Workshops Lecture Notes in Computer Science, 2008, pp. 211– 219.,
doi:10.1007/978-3-540-88875-8_41.
[6] Kumar, S. Gupta, S. K. Singh and K. K. Shukla, "Comparison of various
metrics used in collaborative filtering for recommendation system," 2015

~ 29 ~
Eighth International Conference on Contemporary Computing (IC3),
Noida, 2015, pp. 150-154. Publisher: IEEE.
[7] M. K. Kharita, A. Kumar and P. Singh, "Item-Based Collaborative
Filtering in Movie Recommendation in Real time," 2018 First International
Conference on Secure Cyber Computing and Communication (ICSCCC),
Jalandhar, India, 2018, pp. 340-342. Publisher: IEEE.
[8] Huang, Yao-Chang, and Shyh-Kang Jenor. “An audio recommendation
system based on audio signature description scheme in MPEG-7 Audio.”
2004 IEEE International Conference on Multimedia and Expo (ICME) (IEEE
Cat. No.04TH8763), doi:10.1109/icme.2004.1394273.
[9] . Lund and Y. Ng, "Movie Recommendations Using the Deep Learning
Approach," 2018 IEEE International Conference on Information Reuse and
Integration (IRI), Salt Lake City, UT, 2018, pp. 47-54. Publisher: IEEE.
[10] M. Ilhami and Suharjito, "Film recommendation systems using matrix
factorization and collaborative filtering," 2014 International Conference
on Information Technology Systems and Innovation (ICITSI), Bandung,
2014, pp. 1-6. Publisher: IEEE.

~ 30 ~