Intelligent recommender system based on unsupervised machine learning and demographic attributes

Journal Pre-proof

Intelligent recommender system based on unsupervised machine

learning and demographic attributes

AFOUDI Yassine, LAZAAR Mohamed

PII: S1569-190X(20)30137-4
DOI: https://doi.org/10.1016/j.simpat.2020.102198
Reference: SIMPAT 102198

To appear in: Simulation Modelling Practice and Theory

Received date: 12 May 2020

Revised date: 17 August 2020
Accepted date: 18 September 2020

Please cite this article as: AFOUDI Yassine, LAZAAR Mohamed, Intelligent recommender system
based on unsupervised machine learning and demographic attributes, Simulation Modelling Practice
and Theory (2020), doi: https://doi.org/10.1016/j.simpat.2020.102198

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.

© 2020 Elsevier B.V. All rights reserved.

 Intelligent recommender system based on unsupervised
machine learning and demographic attributes
AFOUDI Yassine, LAZAAR Mohamed
ENSIAS, Mohammed V University in Rabat, Morocco

Abstract
Recommendation systems aim to predict users interests and recommend items most likely
to interest them. In this paper, we propose a new intelligent recommender system that
combines collaborative filtering (CF) with the popular unsupervised machine learning
algorithm K-means clustering. Also, we use certain user demographic attributes such as
the gender and age to create segmented user profiles, when items (movies) are clustered by
genre attributes using K-means and users are classified based on the preference of items
and the genres they prefer to watch. To recommend items to an active user, Collaborative
Filtering approach then is applied to the cluster where the user belongs.
Following the experimentation for well known movies, we show that the proposed system
satisfies the predictability of the CF algorithm in GroupLens. In addition, our proposed
system improves the performance and time response speed of the traditional collaborative
Filtering technique and the Content-Based technique too.

Keywords: Collaborative filtering; K-means clustering; Movies

recommendation; Recommender system

1. Introduction
Today, the internet allows people to access abundant online resources.
Amazon, for example, has a huge collection of products. Although the
amount of information available is increasing day by day, this leaves peo-
ple or specially costumers in new trouble, they find it too difficult to choose
what they really want to see or buy. This is where the recommendation sys-
tem comes in.
Recommendation systems help users find and choose items (books, movies,
restaurants, music... etc.) that may interest them, from the large number of

Email addresses: [email protected] (AFOUDI Yassine),

[email protected] (LAZAAR Mohamed)

Preprint submitted to Simulation Modelling Practice and Theory November 4, 2020

choices available on the web or other electronic information sources. Thats
why we found many companies like Amazon, Netflix and Spotify use rec-
ommendation systems which, of course, allow them to generate significant
engagement and revenue. The reason these companies and others are seeing
increased revenue is that they are bringing real value to their customers, by
suggesting items and making their lives easier, also by knowing what a user
wants the threat of losing a customer to a competitor decreases.
Many approaches are used in recommender systems and we can make a dis-
tinction between them by analyzing the type of data used to generate rec-
ommendations. In general, there are three well-known techniques used in
this field, Collaborative Filtering (CF), Content-Based Filtering (CBF) and
Hybrid models. Collaborative filtering models are based on the idea that
a person likes articles similar to other articles he likes in the past, and the
articles that are liked by other people with similar taste, here we use data
that the user provides, either explicitly (rating) or implicitly (clicking on a
link). As the user provides more inputs or take actions on the suggestions,
the system becomes more and more powerful. Content-Based models are
based on the idea that each product has certain keywords or description,
understands what the user likes based on items he likes before, look up those
keywords or features in the database and recommend different products with
the same attributes. For Hybrid recommender systems combine two or more
recommendation methods to gain better performance with fewer drawbacks.
Most common is collaborative filtering combined with some other techniques.
In this article, we propose a hybrid recommendation system, which combines
collaborative filtering with the K-means unsupervised machine learning algo-
rithm to improve the performance of the traditional CF technique. We first
cluster the items according to some of their features by k-means approach
and after that we analyze the demographic characteristics of users to segment
the users, then classify them according to the upper class of the items that
the user likes. In order to recommend items for a user, we determine a cluster
in advance where the features are similar to the user, then a Collaborative
Filtering applies to the users who belong to the cluster in order to predict
users preference for items.
The rest of the document is organized as follows. The following section briefly
describes the recommendation systems as well as Collaborative Filtering the
best well-known approach and some related work in the context of movie
recommendation. Section 2 presents the proposed approach. Sections 3 eval-
uate our system with other techniques in the area of recommendation and

2
discuss the results. Finally, Section4 summarizes the conclusion.

2. BACKGROUND
2.1. Recommendation Tasks
Recommendation systems have many algorithms that aim to provide the
most relevant and accurate product to the user by filtering useful information
from a huge database pool. In general, recommendation engine based on
three types of approaches collaborative filtering, content based and hybrid
recommender system, but in this article we are interested in Collaborative
Filtering.

2.1.1. Collaborative filtering recommender system

Collaborative filtering is currently one of the most frequently used approaches and gen-
erally provides better results than content-based recommendations, that’s why this method
it can be found in the recommendation systems of Spotify, Netflix and YouTube.There is
a lot of research on collaborative filtering, these types of systems use user interactions
to filter items of interest and the most popular techniques are categorized into two basic
approaches: model-based and memory-based.

2.1.2. Memory-Based Collaborative Filtering

Memory-based collaborative filtering approaches can be divided into two main section’s
user-item filtering and item-item filtering. User-item filtering takes a particular user we
want to give him recommendation items, searches for users similar to this user in another
way users with the same taste based on the similarity of the ratings and recommends items
that those similar users liked. On the other hand, using item-item filtering, we take an
item or product and we search for the users who liked this item, then we find other items
that these users or similar users also liked. It takes the item as input and generates other
items as recommendations.
The basic idea of the two sections is to create a rating matrix, in a simple way, we build
a matrix whose rows represents the users and the columns represent the items and rating
data as values, then we use a similarity approach to find similar users or items, the common
distance metric used in the majority of work is cosine similarity. We can predict user-xs
rating for the movie-y by taking a weighted sum of the movie-y ratings from all other users
where weighting is a similar number between each user and user-x, thats why memory-
based approach is easy to use but its performance decreases when we have sparse data
and it takes a lot of time.

2.1.3. Model-Based Collaborative Filtering

In this approach, CF models are developed using machine learning algorithms and data
mining techniques to predict user rating for items that do not have a rating. The goal is to
train models to be able to make predictions in other words, we extract some information
from the dataset, and use that as a ”model” to make recommendations without needing

3
to use the complete dataset every time. This approach potentially offers the benefits of
both speed and scalability. In this paper, we will use a Model-Based technique based on
matrix factorization named Singular Value decomposition (SVD).

2.2. Related work

Many researches have been done in the field of CF recommendation system and movies
personalization recommendation. Several personalized movie systems have been built in
recent years to help users deal with the large number of existing movies.
Reddy S. et al. [17] propose a method that uses content-based filtering using genre correla-
tion. In Rohan Nayak et al.[14] hybrid system the users will be asked to provide feedback
on some movies and movie genre, and based on the feedback provided, the user will be
segregated, and a set of recommendations will be provided. Katarya Rahul [11] proposes a
hybrid recommender system which utilized k-means clustering algorithm with bio-inspired
artificial bee colony (ABC) optimization technique and applied to the Movielens dataset.
Ponnam, L. T. et al.[15] propose an item based collaborative filtering technique that ex-
amine the user item rating matrix and identify the relationships among various items to
compute the recommendations for the user. In Bagher Rahimpour Cami et al.[3] paper,
they propose a content-based movie recommender system that captures the temporal user
preferences in user modeling and predicts the preferred movies and with the evolution
of deep learning techniques. We found also Rex Ying et al.[22] create an efficient GCN
(Graph Convolutional Network) algorithm for data, which combines efficient random walks
and graph convolutions to generate embeddings incorporations. BogdanWalek et al. [21]
propose a monolithic hybrid recommender system which combines a collaborative filtering
system based on the SVD algorithm with content-based system, and a fuzzy expert system.
The expert system evaluates the importance of the movies based on several parameters
such as the average movie rating, number of ratings, etc. Hong-Quan Do et al. [4] propose
a hybrid recommender system focus on the weighted hybridization and rather than using
fixed weighted for the combination, they aim to offer a simple method to dynamic weight
the combination of CF and CBF. Arisara Pornwattanavichai et al. [16] propose a new
method of recommending Tweets based on hybrid Recommender system with LDA the
unsupervised topic modeling and generalized matrix factorization the supervised learning-
based neural network.
As mentioned above, collaborative filtering is the well-known technique used to give pow-
erful recommendations using the ratings data. Based on this and our selected dataset,
we will continue the research to improve the achievement of results by this technique
by presenting k-means clustering algorithm in the movie recommendation system, which
is based on collaborative filtering recommendation method and demographic attributes
segmentation.

3. PROPOSED WORK ARCHITECTURE

This paper proposes K-means clustering algorithm based on collaborative filtering
approach in the movies recommendation system, Figure 1 represent the proposed archi-
tecture, it mainly consists of the following four functional modules: movies feature extrac-

4
 Figure 1: Adopted architecture

tion module, k-means clustering module, user profile module and finally recommendation
module.

3.1. Movies feature extraction module

We test our system with a movielens dataset, each movie represented with features
such as the title, the movie identifier, and 19 genres attribute (if a movie has a specific
genre we found 1 as a value otherwise 0), for this we plan to use a feature selection or
extraction technique to reduce the number of attributes, in general during our previous
work[2], we found that there was an impact on the use of feature selection techniques on
the results of the recommendations. Principal Component Analysis (PCA) is our choice
to reduce the number of features, this technique is a dimensionality reduction technique
used when we have a slow algorithm because the input dimension is too high and we want
to speed it up.
PCA is a dimensionality reduction that identifies important relationships in our data,
transforms the existing data based on these relationships, and then quantifies the impor-
tance of these relationships so we can keep the most important relationships and delete
the others.
Based on the definition above, we use PCA to reduce the number of features in our movies
database from 19 genres attributes to 10 attributes and by reading the explained variance,
we found that the 10 components contain 86.92% of the item’s information and that’s
good for us, but there is a question of the quality of recommendation using all features
and using only 10 arises. To answer this important question, we test the two datasets
with a combined CF approach with simple k-means, and we evaluate the result using the
RMSE technique (we will explain this technique in the evaluation section), Figure 2 shows
us that using 10 features obtained by PCA gives us much better performance instead of

5
Figure 2: The Collaborative RMSE results using all features and PCA features on 4
random users

using 19 genre features because the Mean error has been reduced. To make the decision
on which datasets we will use, we repeat our test several times, for each one, we found
that the dataset contains PCA features gives us much better performance, that’s why we
will continue our work only with our 10 movie PCA features to minimize the memory
occupied by our system.

3.2. K-means clustering module

This module is used to cluster movies into a specific number of classes, this part
describes the standard k-means algorithm. K-means clustering is one of the simplest and
popular unsupervised machine learning algorithms, as a rule, unsupervised algorithms
make inferences from datasets using only input vectors without referring to known or
labeled results.
Simply, K-means looks for a fixed number (k) of clusters in a dataset, in other words, the
K-means algorithm identifies k number of centroids and calculate the distance between
each object and each cluster center, then assign it to the nearest cluster, update the
averages of all clusters, repeat this process until the criterion function converged.
The K-Means algorithm needs a way to compare the degree of similarity between the
different features. Thus, two data which are similar, will have a reduced dissimilarity
distance, while two different objects will have a greater separation distance, in this work,
we use the k-means clustering algorithm based on the Euclidean similarity approach.
The Euclidean distance is the geometric distance, for example,if we have a matrix X with
i quantitative variables. The Euclidean distance d between two features x1 and x2 is
calculated as follows: v
u n
uX
d(x1 , x2 ) = t (x1n , x2n )2 (1)
i=1

6
After all the information given above, we have arrived at the stage of finding the value of
k. The most common method for choosing the number of clusters is to launch K-Means
with different values of K and calculate the variance of the different clusters. The vari-
ance is the sum of the distances between each centroid of a cluster and the different items
included in the same cluster. Thus, we seek to find a certain number of clusters K so that
the selected clusters minimize the distance between their centroids and the items in the
same cluster. Generally, by plotting the different numbers of clusters K as a function of
the variance, the point of the elbow is that of the number of clusters whose variance no
longer decreases significantly (Elbow approach).
In this work, we give our 10 components from the previous module to the K-means ap-
proach and we obtain as output our dataset of movies classified into a specific number of
clusters.

3.3. User profile module

The main mission of this module is to build for each user a profile based on a set of
parameters for recommendation purposes, in our work we take the rating data from each
user, and we only use movies with a rating greater than or equal to 2.5. After getting the
highest rated movies in another way for an active user, we assign for each movie in the list
its cluster class, after that we calculate the sum of each movie class in the list and we use
the maximum class result as a representative class of the most liked user movie group.
For example, we have a user X who likes movie classes [1,1,1,2,3,3,4,1,1], here, we can
assign that our user liked movies classified in class 1 more than the other classes.

3.4. Recommender system module

This paper proposes a system based on Collaborative Filtering technique and specially
the Model-Based approach, one of the most successful Model-Based CF techniques is
Matrix Factorization. Singular value decomposition (SVD)[19] is the well-known matrix
factorization method and it is our selected method, the general equation of SVD approach
can be expressed as bellow.
X =U ×S×VT (2)
The SVD technique was introduced into the recommendation system domain by Simon
Funk during the Netflix Prize challenge [23].
Based on Linear Algebra, any real matrix R can be decomposed into three matrices U, W,
and V, using this rule SVD result gives three matrices as output U, S and V T , matrix U
represents user vectors and Matrix V T represents movie vectors in our case, when elements
on the diagonal in S are known as singular values of X.
In this step and after classifying our movie datasets into clusters, we take a user and build
a preferred movie class (user profile), then to predict the user ratings of all the movies
using this method, we simply applied the SVD approach with only movies in the preferred
user class and we build a matrix that contains rows representing the users with the same
gender as the active user and columns representing the selected movies and for the values
we use all available ratings, after that we take the dot product of U, S and V T as indicated
in Eq.3. Next, we select the highest rated movies and we give the recommendation after
ignoring the previously viewed movies.
X̂ ≈ U · S · V T (3)

7
4. RESULT AND DISCUSSION
4.1. The dataset
We chose the movielens 100k dataset [9] to evaluate our experiments because it is
widely used and publicly available. MovieLens data sets were collected by the GroupLens
Research Project at the University of Minnesota. We use this dataset for a study where
the goal is to generate recommendation of movies to users. This data set contains 100,000
ratings (1 to 5 scale) from 943 users on 1682 movies and each user have rated at least
20 movies, and there is some demographic information for the users like age, gender,
occupation and zip code.

4.2. Implementation
4.2.1. Experimental steps
First step is to import the dataset in our project, and then we split the whole users
interaction dataset into 75% as the training set and 25% as the test set.
The second step is to use PCA feature extraction to minimize the genre attributes in the
movie dataset from 19 genres to 10 components, after that we use K-means clustering
to group the whole dataset into a specific number of clusters, by plotting the different
10 numbers of clusters as a function of the variance, based on the approach of Elbow
explained above and by observing the Figure 3 we choose the k=6 the number of our
k-means clustering.
After the movies are grouped into 6 clusters,, we assign for each movie its cluster class
number, then we create a user profile for all users according to the preferred viewed class,
then we take an active user and we search for all movies in his favorite class and all users
with his gender, and finally we apply the SVD collaborative filtering approach to make
recommendations and give the result to the user after deleting all the movies previously
viewed.
We will test and evaluate our recommendations by searching whether the recommended
movies to a user according to the training set, are in the item list the user have seen and
rated in the test set.

4.2.2. Experimental platform

All our experiments were implemented using Python and compiled using Jupyter note-
book. We ran all our experiments on a MacBook Pro with Intel core i7 processor having
a speed of 2.7 GHz and 8GB of RAM.

4.2.3. Experiment results

After building our system, we tested it for all users in the dataset and evaluate the
results using time recommendation speed and other offline techniques with the state of
the art approaches in recommendation filed. The result of the experiment is presented in
Table 1 for three random users.

8
 Figure 3: The variance of 10 numbers of clusters on our dataset (Elbow approach)

4.2.4. Evaluation
In the area of recommendation systems, the user wishes to receive the N best recom-
mended items. In this way, the user will view some recommended movies, classified from
best to worst. In fact, in some cases, the user does not care much about the exact order
of the list, a few good recommendations are enough.
As a result, many evaluation techniques are involved in this area, classified as online and
offline techniques. For those online, we need real users to give their opinion and feedback
on the recommendation given, in this work we will use the offline techniques based on the
movies cited for each user in the test set.
In this article, from many offline approaches,we choose to use the RMSE and Precision-
Recall at k techniques. Root Mean Squared Error (RMSE) is a popular technique used to
evaluate a recommender system accuracy based on ratings data. In other way after using
the train set to build our model we predict movies rating for a given user in the test set
and here we compare the predicted ratings with the true values, then we just compute the
square root average of the errors from the whole test set to find the RMSE value as shown
in formula below, when P is the predicted rating and R the true rating.
sP
2
ratings (P − R)
RM SE = (4)
#ratings

The Precision-Recall at k method also is an offline way to evaluate a recommender sys-

tem where k is a definable integer defined by the user to match the goal of the Top-N
recommendations. For this method, we test every model by giving a list of movie recom-
mendations for all users and we evaluate results by calculate and compare the mean of
F-Measure result. Recall and Precision at k are then determined as Equations 5 and 6.

Recall@k = Ri /N r (5)

9
 Table 1: The Top 3 movies recommended with our system on some random users.
User User Preferred User Top three recommended movies
ids movie class Gender
- The Sting
5 class 3 Female - Young Frankenstein
- A Fish Called Wanda
Fargo
17 class 2 Male Twelve Monkeys
Lone Star
- Twelve Monkeys
9 class 2 Male - Fargo
- The Silence of the Lambs

P recision@k = Ri /T r (6)
Where Ri is the number of recommended movies at k that are relevant, Tr is the total of
relevant items and Nr is the number of recommended items at @k. After calculating The
Recall and Precision at k we should normalize the result, then we introduce F-Measure,
which is
F − measure@k = (2 × P @k × R@k)/(P @k + R@k) (7)
After building our system, we want to know and evaluate which model gives us powerful
recommendation and best accuracy of our Movielens 100k database. As explained above
we use precision-recall method, Figure 4 shows us a plot of the mean F-measure at 5 and
10 the number of recommended movies for all users, the Figure 5 shows us the F-measure
at k=5 scatter plot of 10 first users and Figure 6 shows us the Precision curve at k=10 for
first 20 users, when the Figure 7 shows us the time in seconds taken by all the models to
recommend 20 movies to the same three users.
Finally, it should be mentioned that recommendation systems take a bit space and run-
time to give suggestions due to the large number of features and parameters. By observing
and reading the results, we can assume that the use of K-means clustering the popular
unsupervised machine learning technique with Collaborative Filtering associated with de-
mographics attributes gives us much better performance and accuracy, also a fast time
response on movies recommendation than the models based on the traditional Collabora-
tive Filtering or Content-Based approaches but our approach still has a limited point is
the attributes associated with the items. Obviously, if the item does not have descriptive
attributes, it cannot be used in the K-means clustering step to find out which cluster
contains this item.
To show how well our approach is performing, we are testing our model using the IMDB
dataset which contains 671 users with at least 20 interactions and 42,262 unique movies
and 10,0004 user / item interactions, we split the rating data by 75% for the train set and
25% as a test set, reading the result given by the precision recall approach as shown in

10
Figure 4: The mean F-measure plot of six models at K=5 and K=10 using Movielens
dataset

figure 8, we have found that our approach always gives good and better results than the
state of art approaches existing in the field of recommendation. To give more details on
the generosity of our approach, we can say that, if we have an item dataset with many
genre attributes and a user dataset with demographic attributes, our approach can work
well and it will be a good choice to offer customers what they want.
The main Managerial Insights of our approach aim to give a powerful recommendation
in a short time. It also helps to eliminate the limit of the cold start problem and give
suggestions to a new user even if we don’t have much information about user transaction

5. CONCLUSION AND FUTURE WORK

This article proposes an algorithm which uses the unsupervised machine learning al-
gorithm K-means clustering with the collaborative filtering the well-known approach in
the field of recommendation and we take into account certain demographic attributes.
The experimental results show the RMSE was reduced when we use PCA feature extrac-
tion instead of using all 19 movie features, after that we take those features for building
our system based on clustering movies into 6 clusters, then we take the most viewed movie
class for an active user and all users with the same gender and we apply the SVD collab-
orative filtering to give powerful recommendations. By using Precision-Recall the offline
approach for evaluating a recommender system we found that the proposed algorithm can
efficiently improve the accuracy and the performance of movies recommendation. So the
proposed method is feasible.
In another study, we will continue to explore and improve other techniques used in
recommendation systems and try to use advanced machine learning algorithms and deep
learning models.

11
 Figure 5: The F-measure scatter plot of 10 users

Figure 6: The Precision curve at k=10 for first 20 users

12
Figure 7: The time in seconds taken by all the models to recommend movies to three users

Figure 8: Global F-measure plot at K=5 and K=10 for 671 users using IMDB dataset

13
ACKNOWLEDGMENT
The authors would like to thank the Smart System Lab our research laboratory, and
Al Borchers for cleaning up this data.

References
[1] Afoudi, Y., Lazaar, M., Al Achhab, M., 2019. Collaborative Filtering Recommender
System. In: Advanced Intelligent Systems for Sustainable Development (AI2SD2018).
Advances in Intelligent Systems and Computing. Springer International Publishing,
Cham, pp. 332–345.

[2] Afoudi, Y., Lazaar, M., Al Achhab, M., 2019. Impact of Feature selection on content-
based recommendation system. In: 2019 International Conference on Wireless Tech-
nologies, Embedded and Intelligent Systems (WITS). pp. 1–6.

[3] Cami, B. R., Hassanpour, H., Mashayekhi, H., 2017. A content-based movie recom-
mender system based on temporal user preferences. In: 2017 3rd Iranian Conference
on Intelligent Systems and Signal Processing (ICSPIS). IEEE, Shahrood, pp. 121–125.

[4] Do, H.-Q., Le, T.-H., Yoon, B., ???? Dynamic weighted hybrid recommender systems.
In: 2020 22nd International Conference on Advanced Communication Technology
(ICACT). pp. 644–650.

[5] Du, Z., Zhang, T., Chen, Y., Ai, L., Wang, X., 2011. A content and user-oblivious
video-recommendation algorithm. Simulation Modelling Practice and Theory 19 (9),
1895–1912.

[6] Duwairi, R., Abu-Rahmeh, M., 2015. A novel approach for initializing the spherical
K-means clustering algorithm. Simulation Modelling Practice and Theory 54, 49–63.

[7] Duwairi, R., Ammari, H., 2016. An enhanced CBAR algorithm for improving recom-
mendation systems accuracy. Simulation Modelling Practice and Theory 60, 54–68.

[8] Graef, G., Schaefer, C., 2002. Application of ART2 Networks and Self-Organizing
Maps to Collaborative Filtering. In: Hypermedia: Openness, Structural Awareness,
and Adaptivity. Lecture Notes in Computer Science. Springer, Berlin, Heidelberg,
pp. 296–309.

[9] Harper, F. M., Konstan, J. A., 2016. The MovieLens Datasets: History and Context.
ACM Transactions on Interactive Intelligent Systems 5 (4), 1–19.

[10] Jakomin, M., Curk, T., Bosni, Z., 2018. Generating inter-dependent data streams for
recommender systems. Simulation Modelling Practice and Theory 88, 1–16.

[11] Katarya, R., 2018. Movie recommender system with metaheuristic artificial bee. Neu-
ral Computing and Applications 30 (6), 1983–1990.

14
[12] Lee, M., Choi, P., Woo, Y., 2002. A Hybrid Recommender System Combining Col-
laborative Filtering with Neural Network. In: Adaptive Hypermedia and Adaptive
Web-Based Systems. Lecture Notes in Computer Science. Springer, Berlin, Heidel-
berg, pp. 531–534.

[13] Ma, Z., Yang, Y., Wang, F., Li, C., Li, L., 2014. The SOM Based Improved K-Means
Clustering Collaborative Filtering Algorithm in TV Recommendation System. In:
2014 Second International Conference on Advanced Cloud and Big Data. pp. 288–
295.

[14] Nayak, R., Mirajkar, A., Rokade, J., Wadhwa, G., 2018. Hybrid Recommendation
System For Movies 05 (03), 4.

[15] Ponnam, L. T., Deepak Punyasamudram, S., Nallagulla, S. N., Yellamati, S., 2016.
Movie recommender system using item based collaborative filtering technique. In:
2016 International Conference on Emerging Trends in Engineering, Technology and
Science (ICETETS). IEEE, Pudukkottai, India, pp. 1–5.

[16] Pornwattanavichai, A., sakolnagara, P. B. n., Jirachanchaisiri, P., Kitsupapaisan, J.,

Maneeroj, S., ???? Enhanced tweet hybrid recommender system using unsupervised
topic modeling and matrix factorization-based neural network. In: Supervised and
Unsupervised Learning for Data Science. Springer International Publishing, pp. 121–
143.

[17] Reddy, S., Nalluri, S., Kunisetti, S., Ashok, S., Venkatesh, B., 2019. Content-Based
Movie Recommendation System Using Genre Correlation. In: Smart Intelligent Com-
puting and Applications. Smart Innovation, Systems and Technologies. Springer, Sin-
gapore, pp. 391–397.

[18] Vembu, S., Baumann, S., 2005. A Self-Organizing Map Based Knowledge Discovery
for Music Recommendation Systems. In: Computer Music Modeling and Retrieval.
Lecture Notes in Computer Science. Springer, Berlin, Heidelberg, pp. 119–129.

[19] Vozalis, M. G., Margaritis, K. G., 2007. Using svd and demographic data for the
enhancement of generalized collaborative filtering. Information Sciences 177 (15),
3017–3037.

[20] Wadia, K., Gupta, P., 2011. Movie recommendation system based on self-organizing
maps. The University of Texas at Austin, Austin, Texas.

[21] Walek, B., Fojtik, V., 2020. A hybrid recommender system for recommending relevant
movies using an expert system 158, 113452.

[22] Ying, R., He, R., Chen, K., Eksombatchai, P., Hamilton, W. L., Leskovec, J., 2018.
Graph Convolutional Neural Networks for Web-Scale Recommender Systems. In: Pro-
ceedings of the 24th ACM SIGKDD International Conference on Knowledge Discov-
ery & Data Mining. KDD ’18. Association for Computing Machinery, London, United
Kingdom, pp. 974–983.

15
[23] Zhou, Y., Wilkinson, D., Schreiber, R., Pan, R., 2008. Large-Scale Parallel Collab-
orative Filtering for the Netflix Prize. In: Algorithmic Aspects in Information and
Management. Lecture Notes in Computer Science. Springer, Berlin, Heidelberg, pp.
337–348.

16