ABSTRACT

1. This paper reviews the research trends that link the advanced technical aspects of
recommendation systems that are used in various service areas and the business aspects of
these services. First, for a reliable analysis of recommendation models for recommendation
systems, data mining technology, and related research by application service, more than
135 top-ranking articles and top-tier conferences published in Google Scholar between 2010
and 2021 were collected and reviewed. Based on this, studies on recommendation system
models and the technology used in recommendation systems were systematized, and
research trends by year were analyzed. In addition, the application service fields where
recommendation systems were used were classified, and research on the recommendation
system model and recommendation technique used in each field was analyzed.
Furthermore, vast amounts of application service-related data used by recommendation
systems were collected from 2010 to 2021 without taking the journal ranking into
consideration and reviewed along with various recommendation system studies, as well as
applied service field industry data. As a result of this study, it was found that the flow and
quantitative growth of various detailed studies of recommendation systems interact with the
business growth of the actual applied service field. While providing a comprehensive
summary of recommendation systems, this study provides insight to many researchers
interested in recommendation systems through the analysis of its various technologies and
trends in the service field to which recommendation systems are applied.

2. Recommender systems are efficient tools for filtering online information, which is
widespread owing to the changing habits of computer users, personalization trends, and
emerging access to the internet. Even though the recent recommender systems are eminent
in giving precise recommendations, they suffer from various limitations and challenges like
scalability, cold-start, sparsity, etc. Due to the existence of various techniques, the selection
of techniques becomes a complex work while building application-focused recommender
systems. In addition, each technique comes with its own set of features, advantages and
disadvantages which raises even more questions, which should be addressed. This paper
aims to undergo a systematic review on various recent contributions in the domain of
recommender systems, focusing on diverse applications like books, movies, products, etc.
Initially, the various applications of each recommender system are analysed. Then, the
algorithmic analysis on various recommender systems is performed and a taxonomy is
framed that accounts for various components required for developing an effective
recommender system. In addition, the datasets gathered, simulation platform, and
performance metrics focused on each contribution are evaluated and noted. Finally, this
review provides a much-needed overview of the current state of research in this field and
points out the existing gaps and challenges to help posterity in developing an efficient
recommender system.

3. On the Internet, where the number of choices is overwhelming, there is need to

filter, prioritize and efficiently deliver relevant information in order to alleviate the problem
of information overload, which has created a potential problem to many Internet
users. Recommender systems solve this problem by searching through large volume of
dynamically generated information to provide users with personalized content and services.
This paper explores the different characteristics and potentials of different prediction
techniques in recommendation systems in order to serve as a compass for research and
practice in the field of recommendation systems.

4. Recommender systems are similar to an information filtering system that helps

identify items that best satisfy the users’ demands based on their preference profiles.
Context-aware recommender systems (CARSs) and multi-criteria recommender systems
(MCRSs) are extensions of traditional recommender systems. CARSs have integrated
additional contextual information such as time, place, and so on for providing better
recommendations. However, the majority of CARSs use ratings as a unique criterion for
building communities. Meanwhile, MCRSs utilize user preferences in multiple criteria to
better generate recommendations. Up to now, how to exploit context in MCRSs is still an
open issue. This paper proposes a novel approach, which relies on deep learning for context-
aware multi-criteria recommender systems. We apply deep neural network (DNN) models to
predict the context-aware multi-criteria ratings and learn the aggregation function. We
conduct experiments to evaluate the effect of this approach on the real-world dataset. A
significant result is that our method outperforms other state-of-the-art methods for
recommendation effectiveness.

5. The last decade has witnessed the generation of an overwhelming amount of

information online. Such a high volume of information makes it difficult for users to find their
desired content (movies, music, books, etc.) in a reasonable time. There is need to filter and
efficiently deliver relevant information to alleviate the information overload problem, which
has created a potential issue to many Internet users. One way of solving this problem is by
using recommender systems. Recommender Systems (RS) are information filtering systems
that cope with the information overload problem by filtering vital information fragment out
of large amount of dynamically generated information according to user’s preferences about
items. RS became an important area of research, thanks to their help for users to suggest
the items they might prefer, instead of crawling thousands or hundreds of items until finding
the most adequate. However, there are usually various factors that may impact users
preferences. Therefore, research in recommender systems is starting to recognise the
importance of items multi-criteria preferences and the role of user’s context in enhancing
the recommendation output. In this respect, traditional recommender systems are extended
to offer novel lines of research areas such as Context-Aware Recommendation Systems
(CARS) and Multi-Criteria Recommendation Systems (MCRS). This thesis investigates the
inclusion of useful additional information in the recommendation process. Firstly, two novel
collaborative filtering based approaches for context-aware recommendation are proposed.
The first approach is based on a neighborhood-based model integrating user’s inferred
contextual situation in the rating prediction computing process. The second one presents a
matrix factorizationbased model that consists of two strategies: a weighting strategy that
integrates the relevant contextual dimensions weights in the rating prediction process and
an interaction strategy that incorporates the interaction measurements between correlated
contextual dimensions in the rating prediction process. Despite much of work has been done
on extended recommender systems, the interesting research direction including both
context-awareness and multi-criteria directions remain unexplored, where these directions
are addressed separately in most existent literature. Therefore, we aim to capture more fine
grained preferences to upgrade items recommendation quality by integrating users multi-
criteria preferences under specific contexts. With this aim in mind, two new context-aware
multi-criteria recommendation models are proposed. The first model focuses on estimating
users overall ratings through predicting clustered criteria ratings then using prioritized
aggregation operators as means of multi-criteria ratings aggregation. The second one
addresses the prediction of users preferences through predicting clustered criteria ratings
by considering the dependencies between users and contexts as well as the dependencies
between the criteria. The validation of the proposed recommendation approaches are
conducted using realworld datasets, novel created datasets and popular metrics. The
obtained results demonstrate that our proposals exhibit significant improvements over
alternative recommendation approaches.

6. In e-commerce today, contents available for users to explore are overwhelming

because an average ecommerce website is about seventy per cent (70%) more than a
physical store in total number of users and items. Hence, the need to filter, prioritize and
efficiently deliver relevant information using recommender systems. We will design and
develop a recommendation model that uses object-oriented analysis and design
methodology (OOADM), improved collaborative filtering algorithm and an efficient quick sort
algorithm to solve these problems. This will be achieved by implementing the stated model
with python model-view-controller (MVC) framework known as Django Framework. This
improved system is implemented using a real-time, cloud-hosted NOSQL database called
FireBase which guarantees scalability. From the results, the speed and scalability of book
recommendation was improved with a performance record obtained within the range of
ninety (90) to ninety-five (95) per cent using the root mean square error (RMSE) of several
recommendations obtained from the system.

7. The aim of this project is to employ collaborative filtering to create a

recommendation system. This will be achieved through the utilization of the Python
programming language and the scikitsurprise library. The dataset used in this project is the
Movie Lens dataset, which contains information about movies and ratings provided by users.
The project involves creating an SVD model with 50 latent factors, fitting the model with the
training data, and using the model to predict ratings for non-interacted movies. The
suggested items are arranged in a decreasing sequence based on the anticipated ratings,
while the model's effectiveness is assessed by computing the RMSE.. The results show that
the SVD model with 50 latent factors performs well in predicting ratings for noninteracted
movies. The top recommended movies for user 1 have predicted ratings ranging from 4.8 to
4.2. The RMSE value obtained is 0.82, which indicates that the model's predictions are
accurate. Overall, the project demonstrates the effectiveness of collaborative filtering in
creating personalized recommendations for users. The implementation can be further
improved by incorporating other features such as user demographics, movie genres, and
contextual information.

8. Te explosive growth in the amount of available digital information in higher

education has created a potential challenge of information overload, which hampers timely
access to items of interest. Te recommender systems are applied in diferent domains such
as recommendations flm, tourist advising, webpages, news, songs, and products. But the
recommender systems pay less attention to university library services. Te most users of
university library are students. Tese users have a lack of ability to search and select the
appropriate materials from the large repository that meet for their needs. A lot of work has
been done on recommender system, but there are technical gaps observed in existing works
such as the problem of constant item list in using web usage mining, decision tree induction,
and association rule mining. Besides, it is observed that there is cold start problem in case-
based reasoning approach. Terefore, this research work presents matrix factorization
collaborative fltering with some performance enhancement to overcome cold start problem.
In addition, it presents a comparative study among memory-based and model-based
approaches. In this study, researchers used design science research method. Te study
dataset, 5189 records and 76,888 ratings, was collected from the University of Gondar
student information system and online catalogue system. To develop the proposed model,
memory-based and model-based approaches have been tested. In memory-based approach,
matrix factorization collaborative fltering with some performance enhancements has been
implemented. In model-based approach, Knearest neighbour (KNN) and singular value
decomposition (SVD) algorithms are also assessed experimentally. Te SVD model is trained
on our dataset optimized with a scored RMSE 0.1623 compared to RMSE 0.1991 before the
optimization. Te RMSE for a KNN model trained using the same dataset was 1.0535. Tis
indicates that the matrix factorization performs better than KNN models in building
collaborative fltering recommenders. Te proposed SVD-based model accuracy score is 85%.
Te accuracy score of KNN model is 53%. So, the comparative study indicates that matrix
factorization technique, specifcally SVD algorithm, outperforms over neighbourhood-based
recommenders. Moreover, using hyperparameter tuning with SVD also has an improvement
on model performance compared with the existing SVD algorithm.

9. In the past decade, Machine Learning (ML) models have become a critical part of
large scale analytics frameworks to solve different problems, such as identify trends and
patterns in the data, manipulate images, classify text, and produce recommendations. For
the latter (i.e., produce recommendations), ML frameworks have been extended to
incorporate both specific recommendation algorithms (e.g., SlopeOne [1]), but also more
generalised models (e.g., K-Nearest Neighbours (KNN) [2]) that can be applied not only to
recommendation tasks, such as rating prediction or item ranking, but also other classes of
ML problems. This thesis examines an important and popular area of the Recommendation
Systems (RS) design space, focusing on algorithms that are both specifically designed for
producing recommendations, as well as other types of algorithms that are also found in the
wider ML field. However, the latter will be only showcased in RS-based use-cases to allow
comparison with specific RS models. Throughout the past years, there have been increased
interest in RS from both academia and industry, which led to the development of numerous
recommendation algorithms [3]. While there are different families of recommendation
models (e.g., Matrix Factorisation (MF)- based, K-Nearest Neighbours (KNN)-based), they can
be grouped in three classes as follows: Collaborative Filtering (CF), Content-based Filtering
(CBF), and Hybrid Approaches (HA). This thesis investigates the most popular class of RS,
namely Collaborative Filtering-based (CF) recommendation algorithms, which recommend
items to a user based on similar users’ preferences. One of the current challenges in
building CF engines is the selection of the algorithms to be used for producing
recommendations. It is often the case that a one-CFmodel-fits-all solution becomes
unfeasible due to the dynamic relationship between users and items, and the rate at which
new algorithms are proposed in the literature. This challenge is exacerbated by the constant
growth of the input data, which in turn impacts the efficiency of these models, as more
computational resources are required to train the algorithms on large collections to attain a
predefined/desired quality of recommendations. In CF, these challenges have also impacted
the way providers deliver content to the users, as they need to strike a balance between
revenue maximisation (i.e., how many resources are spent for training the CF models) and
the users’ satisfaction (i.e., produce relevant recommendations for the users). In addition, CF
models need to be periodically retrained to capture the latest user preferences and
interactions with the items, and hence, content providers have to decide whether and when
to retrain their CF algorithms, such that the high training times and resource utilisation costs
are kept within the operational and monetary budget. Therefore, the problem of estimating
resource consumption for CF becomes of critical importance. In this thesis, we address the
pressing challenge of predicting the efficiency (i.e., computational resources spent during
training) of traditional and neural CF for a number of popular representatives, including
algorithms based on Matrix Factorisation (MF), KNearest Neighbours (KNN), Co-clustering,
Slope One schemes, as well as well-known types of Deep Learning (DL) architectures, such
as Variational Autoencoder (VAE), Multilayer Perceptron (MLP), and Convolutional Neural
Network (CNN). To this end, we first study the computational complexity of the training
phase of said CF models and derive time and space complexity equations. Then, using
characteristics of the input and the aforementioned equations, we contribute a methodology
for predicting the processing time, memory overhead, and GPU utilisation of the CF’s
training phase. Our contributions further include an adaptive sampling strategy, to address
the trade-off between the computational cost of sampling the dataset and training the CF
models on the said samples and the accuracy of the estimated resource consumption of the
CF trained on a full collection. Furthermore, we provide a framework which quantifies both
the training efficiency (i.e., resource consumption) of CF, as well as the quality of the
recommendations produced by the said CF once it has been trained. Finally, systematic
experimental evaluations demonstrate that our methodology outperforms state-of-the-art
regression schemes (i.e., BB/GBM) by a considerable margin (e.g., for predicting the
processing time of CF, the accuracy of WB/LR is 160% higher than the one of BB/GBM), with
an overhead that is a small fraction (e.g., 3-4 times smaller) of the overall requirements of
CF training.

10. Recommender systems have been an efcient strategy to deal with information
overload by producing personalized predictions. Recommendation systems based on deep
learning have accomplished magnifcent results, but most of these systems are traditional
recommender systems that use a single rating. In this work, we introduce a multi-criteria
collaborative fltering recommender by combining deep neural network and matrix
factorization. Our model consists of two parts: the frst part uses a fused model of deep
neural network and matrix factorization to predict the criteria ratings and the second one
employs a deep neural network to predict the overall rating. The experimental results on
two datasets, including a real-world dataset, show that the proposed model outperformed
several state-of-the-art methods across diferent datasets and performance evaluation
metrics.

11. Numerous research studies have emphasized the significance of contextual

information when it comes to recommender models. This importance is especially evident in
the realm of e-commerce platforms, where recommender systems have been effectively
suggesting products and services to users by integrating contextual data into their models.
By doing so, these systems can better understand user preferences and behaviors during
transactions on the platform. As a result, a growing number of platforms are now collecting
evaluation values for products and services based on various user contexts, leading to the
emergence of multi-context-based rating datasets. This presents a valuable opportunity to
implement multi-criteria collaborative filtering models, which we propose as a solution. Our
approach involves integrating user contextual rating data and conducting experiments using
two sets of contextual evaluation datasets: De Paul Movie and In Car Music. The results
demonstrate that the multi-criteria collaborative filtering model outperforms the single-
contextbased collaborative filtering model in terms of accuracy. This study opens up
promising avenues for future research aimed at further enhancing recommendation
accuracy for customers on online sales platforms.

12.The importance of contextual information has been recognized by researchers

and practitioners in many disciplines, including e-commerce personalization, information
retrieval, ubiquitous and mobile computing, data mining, marketing, and management.
While a substantial amount of research has already been performed in the area of
recommender systems, most existing approaches focus on recommending the most relevant
items to users without taking into account any additional contextual information, such as
time, location, or the company of other people (e.g., for watching movies or dining out). In
this chapter we argue that relevant contextual information does matter in recommender
systems and that it is important to take this information into account when providing
recommendations. We discuss the general notion of context and how it can be modeled in
recommender systems. Furthermore, we introduce three different algorithmic paradigms –
contextual prefiltering, post-filtering, and modeling – for incorporating contextual
information into the recommendation process, discuss the possibilities of combining several
contextaware recommendation techniques into a single unifying approach, and provide a
case study of one such combined approach. Finally, we present additional capabilities for
context-aware recommenders and discuss important and promising directions for future
research.

13.With the increasing volume of news articles available on the internet, personalized
news recommendations have become increasingly important for users to discover relevant
and interesting news articles. However, traditional recommender systems often fail to
capture the dynamic nature of users' preferences and the changing trends in news articles.
To address this challenge, this paper proposes a context-aware personalized news
recommendation system that incorporates contextual information to enhance the
personalization of news recommendations. The approach involves collecting, extracting,
exploring, cleaning, and processing a large dataset of news articles from 19 distinct internet
news sources, totaling 22,657 English pieces. Four diferent recommender systems were built
using diferent techniques, including content-based methods such as TF-IDF, Bag-of-Words,
and Word2Vec, and a collaborative fltering system based on click behavior. To evaluate the
efectiveness of our models, we used a combination of standard comparison metric, including
precision and recall, and user feedback. To demonstrate the applicability of the model, a
web interface was constructed, and we used RMSE and MAE to evaluate the performance of
the collaborative fltering model. In addition, we conducted a comparative study to compare
the accuracy of diferent algorithms with diferent baseline methods, including random and
recency. The evaluation results showed that incorporating contextual information and
collaborative fltering can signifcantly improve the personalization of news
recommendations. The study suggests that the collaborative fltering model based on click
behavior is the most efective approach, with a mean MAE of 0.0252 and a mean RMSE of
0.0364. The content-based models were also efective approaches for recommending news
articles, outperforming the baseline approaches.

14. In an era where digital information is abundant, the role of recommender systems
in navigating this vast landscape has become increasingly vital. This study proposes a novel
deep learning-based approach integrating multi-context and multi-criteria data within a
unified neural network framework. The model processes these dimensions concurrently,
significantly improving the precision of personalized recommendations. Traditionally, such
systems use a two-dimensional user-item preference matrix based on users’ overall ratings.
Context-aware and multi-criteria recommender systems extend traditional methods with
context awareness and multiple criteria. In contrast to traditional methods, our approach
intricately weaves together multi-context and multi-criteria data within its architecture. This
concurrent processing enables sophisticated interactions between context and criteria,
enhancing recommendation accuracy. While context-aware systems incorporate contextual
information such as time and location when making recommendations, multicriteria-based
approaches offer a spectrum of evaluative criteria, enriching the user experience with more
tailored and relevant suggestions. Although both approaches have advantages in producing
more accurate and personalized referrals, context information and multi-criteria ratings
have not been employed together for producing recommendations. Our research proposes a
novel deep learning-based approach for the multicontext, multi-criteria recommender
system to address this gap. In contrast to traditional approaches that process context-aware
recommender systems and multi-criteria recommender systems separately, our deep
learning model intricately weaves together multi-context and multi-criteria data within its
architecture. This integration is not staged; both dimensions are concurrently processed
through a unified neural network framework. The model facilitates a sophisticated
interaction between context and criteria by embedding these elements into the core of the
network’s multiple layers. This methodology enhances the system’s adaptability and
significantly improves its precision in delivering personalized recommendations, leveraging
the compounded effects of contextual and criteria-specific insights. Experiments on
TripAdvisor and ITMRec multi-context multi-criteria datasets demonstrate that the proposed
method provides more accurate predictions than the state-of-the-art recommendation
techniques. Context-aware multi-criteria ratings offer a promising opportunity to improve
recommendation models.

15. Traditional collaborative filtering based recommender systems deal with the two-
dimensional user-item rating matrix where users have rated a set of items into the system.
Although traditional recommender systems are widely adopted but they are unable to
generate effective recommendations in case of multi-dimensionality i.e. multi-criteria
ratings, contextual information, side information etc. The curse of dimensionality is the
major issue in the recommendation systems. Therefore, in this paper, we proposed a
clustering approach to incorporate multi-criteria ratings into traditional recommender
systems effectively. Furthermore, we compute the intra-cluster user similarities using a
Mahalanobis distance method in order to make more accurate recommendations and
compared the proposed approach with the traditional collaborative filtering based method
using Yahoo! Movies dataset

In recent years, recommender systems (RS) have become essential tools for
filtering overwhelming volumes of information across domains such as e-commerce,
entertainment, and education. These systems aim to deliver personalized
recommendations based on user preferences, but they face challenges like
scalability, cold-start issues, and data sparsity. Traditional collaborative filtering
(CF) methods, while widely adopted, often fall short when handling complex, multi-
dimensional data such as contextual information and multi-criteria ratings. To
address these limitations, research has advanced towards context-aware
recommender systems (CARS) and multi-criteria recommender systems (MCRS),
integrating additional data such as time, location, and user-specific criteria into the
recommendation process.

This thesis explores novel approaches to enhance recommendation accuracy by

combining context-aware and multi-criteria collaborative filtering methods. Using
machine learning models, such as deep neural networks and matrix factorization,
this work proposes two hybrid models that integrate users' contextual situations
and multi-criteria ratings to improve recommendation performance. Experimental
results, validated on real-world datasets, demonstrate that the proposed methods
outperform traditional single-criterion systems in terms of precision and recall
metrics.

By systematically analyzing recommendation models and applying advanced

algorithms, this research contributes to filling the gap between theory and practice
in multi-dimensional RS. The findings highlight the potential for improved
personalization and efficiency, particularly in dynamically generated content
environments, where user preferences evolve rapidly. The proposed models not
only enhance recommendation accuracy but also pave the way for future work in
extending recommender systems to incorporate even more complex user-data
relationships.