Imen Boukhris | Université de la Manouba, Tunis (original) (raw)

Papers by Imen Boukhris

Research paper thumbnail of A GAN-BERT based decision making approach in peer review

Social network analysis and mining, May 22, 2024

Research paper thumbnail of GAN Based Restyling of Arabic Handwritten Historical Documents

Research paper thumbnail of Towards more trustworthy predictions: A hybrid evidential movie recommender system

JUCS - Journal of Universal Computer Science

Recommender Systems (RSs) are considered as popular tools that have revolutionized the e-commerce... more Recommender Systems (RSs) are considered as popular tools that have revolutionized the e-commerce and digital marketing. Their main goal is predicting the users’ future preferences and providing accessible and personalized recommendations. However, uncertainty can spread at any level throughout the recommendation process, which may affect the results. In fact, the ratings given by the users are often unreliable. The final provided predictions itself may also be pervaded with uncertainty and doubt. Obviously, the reliability of the predictions cannot be fully certain and trustworthy. For the system to be effective, recommendations must inspire trust in the system and provide reliable and credible recommendations. The user may speculate about the uncertainty pervaded behind the given recommendation. He could tend to a reliable recommendation offering him a global overview about his preferences rather than an inappropriate one that contradicts his activities and objectives. While such ...

Research paper thumbnail of A new evidential collaborative filtering: A hybrid memory- and model-based approach

Data Science and Knowledge Engineering for Sensing Decision Support, 2018

Research paper thumbnail of A worker clustering-based approach of label aggregation under the belief function theory

Applied Intelligence, 2018

Crowdsourcing platforms have been attracting a wide attention in the field of artificial intellig... more Crowdsourcing platforms have been attracting a wide attention in the field of artificial intelligence in recent years, providing a cheap and reachable human-powered resource to gather massive labeled data. These data are used to effectively build supervised learning models for academic research puposes. However, despite the attractiveness of these systems, the major concern has always been the quality of the collected labels. In fact, a wide range of workers contributes in labeling data leading to be in possession of potentially noisy and imperfect labels. Therefore in this paper, we propose a new label aggregation technique that allows to determine workers qualities via a clustering process and then represent and combine their labels to estimate the final one under the belief function theory. This latter is notorious for its strength and flexibility when dealing with imperfect information. Experimental results demonstrate that our proposed method outperforms the related work baseline and improves results quality.

Research paper thumbnail of Evidential Item-Based Collaborative Filtering

Lecture Notes in Computer Science, 2016

Recommender Systems (RSs) in particular the collaborative filtering approaches have reached a hig... more Recommender Systems (RSs) in particular the collaborative filtering approaches have reached a high level of popularity. These approaches are designed for predicting the user’s future interests towards unrated items. However, the provided predictions should be taken with restraint because of the uncertainty pervading the real-world problems. Indeed, to not give consideration to such uncertainty may lead to unrepresentative results which can deeply affect the predictions’ accuracy as well as the user’s confidence towards the RS. In order to tackle this issue, we propose in this paper a new evidential item-based collaborative filtering approach. In our approach, we involve the belief function theory tools as well as the Evidential K-Nearest Neighbors (EKNN) classifier to deal with the uncertain aspect of items’ recommendation ignored by the classical methods. The performance of our new recommendation approach is proved through a comparative evaluation with several traditional collaborative filtering recommenders.

Research paper thumbnail of An Evidential Semi-supervised Label Aggregation Approach

Crowdsourcing is a powerful concept that typically takes advantage of human intelligence to deal ... more Crowdsourcing is a powerful concept that typically takes advantage of human intelligence to deal with problems in many fields most importantly in machine learning. Indeed, it enables to collect training labels in a fast and cheap way for supervised algorithms. The only major challenge is that the quality of the contributions is not always guaranteed because of the expertise heterogeneity of the participants. One of the basic strategies to overcome this problem is to assign each task to multiple workers and then combine their answers in order to obtain a single reliable one. This paper provides a new iterative approach that aggregates imperfect labels using the supervision of few gold labels under the evidence theory. Besides of inferring the consensus answers, the workers’ accuracies and the questions difficulties are as well estimated. A comparative evaluation on synthetic and real datasets confirms the effectiveness of our semi-supervised approach over the baselines.

Research paper thumbnail of An Adaptive Approach of Label Aggregation Using a Belief Function Framework

Lecture Notes in Business Information Processing, 2017

Crowdsourcing knows a large expansion in recent years. It is widely used as a low-cost alternativ... more Crowdsourcing knows a large expansion in recent years. It is widely used as a low-cost alternative to guess the true labels of training data in machine learning problems. In fact, crowdsourcing platforms such as Amazon’s Mechanical Turk allow to collect from crowd workers multiple labels aggregated thereafter to infer the true label. As the workers are not always reliable, imperfect labels can occur. In this work, we propose an approach that aggregates labels using the belief function theory besides of adaptively integrating both labelers expertise and question difficulty. Experiments with real data demonstrate that our method provides better aggregation results.

Research paper thumbnail of Academic Venue Recommendation Based on Refined Cross Domain

Intelligent Systems Design and Applications, 2022

Research paper thumbnail of Big Data Classification Using Belief Decision Trees: Application to Intrusion Detection

Advances in Intelligent Systems and Computing, 2015

Over the past few years, the data volume explosion fueled by exciting progression in computer tec... more Over the past few years, the data volume explosion fueled by exciting progression in computer technologies, made the Big Data the focus of widespread attention. Big Data is nebulous since it is an interaction result of several dimensions of scale, among them the veracity which refers to biases and noise in data. Therefore, Big Data veracity is a challenge because it requires a different approach in order to cope with this imperfection. We propose to involve the belief function theory and the belief decison tree as a classification technique to accommodate large applications where the uncertainty reigns. In this paper, we will be firstly concerned with the construction of the belief decision tree, using MapReduce programming model and the averaging approach as a classification method under uncertainty. Then, we will conduct experiments on intrusion detection massive data set, to distinguish between attacks and normal connections in such uncertain context.

Research paper thumbnail of A New User-Based Collaborative Filtering Under the Belief Function Theory

Advances in Artificial Intelligence: From Theory to Practice, 2017

The collaborative filtering (CF) is considered as the most widely used approach in the field of R... more The collaborative filtering (CF) is considered as the most widely used approach in the field of Recommender Systems (RSs). It tends to predict the users’ preferences based on the users sharing similar interests. However, ignoring the uncertainty involved in the provided predictions is among the limitations related to this approach. To deal with this issue, we propose in this paper a new user-based collaborative filtering within the belief function theory. In our approach, the evidence of each similar user is taken into account and Dempster’s rule of combination is used for combining these pieces of evidence. A comparative evaluation on a real world data set shows that the proposed method outperforms traditional user-based collaborative filtering recommenders.

Research paper thumbnail of Iterative Aggregation of Crowdsourced Tasks Within the Belief Function Theory

Lecture Notes in Computer Science, 2017

With the growing of crowdsourcing services, gathering training data for supervised machine learni... more With the growing of crowdsourcing services, gathering training data for supervised machine learning has become cheaper and faster than engaging experts. However, the quality of the crowd-generated labels remains an open issue. This is basically due to the wide ranging expertise levels of the participants in the labeling process. In this paper, we present an iterative approach of label aggregation based on the belief function theory that simultanously estimates labels, the reliability of participants and difficulty of each task. Our empirical evaluation demonstrate the efficiency of our method as it gives better quality labels.

Research paper thumbnail of An Evidential Imprecise Answer Aggregation Approach Based on Worker Clustering

Intelligent Data Engineering and Automated Learning – IDEAL 2019, 2019

Crowdsourcing has become a popular and practical tool to gather low-cost labels from human worker... more Crowdsourcing has become a popular and practical tool to gather low-cost labels from human workers in order to provide training data for machine learning applications. However, the quality of the crowdsourced data has always been an issue mainly caused by the quality of the contributors. Since they can be unreliable due to many factors, it became common to assign a task to more than one person and then combine the gathered contributions in order to obtain high quality results. In this work, we propose a new approach of answer combination within an evidential framework to cope with uncertainty. In fact, we assume that answers could be partial which means imprecise or even incomplete. Moreover, the approach includes an important step that clusters workers using the k-means algorithm to determine their types in order to effectively integrate them in the aggregation of answers step. Experimentation on simulated dataset show the efficiency of our approach to improve outcome quality.

Research paper thumbnail of An Evidential Collaborative Filtering Approach Based on Items Contents Clustering

Belief Functions: Theory and Applications, 2018

Research paper thumbnail of Exploring Location and Ranking for Academic Venue Recommendation

Advances in Intelligent Systems and Computing, 2018

Publishing scientific results is extremely important for each researcher. The concrete challenge ... more Publishing scientific results is extremely important for each researcher. The concrete challenge is how to select the right academic venue that corresponds to researcher's current interest and without missing the deadline at the same time. Due to the huge number of academic venues especially in the field of computer science, it is difficult for researchers to choose a conference or a journal to submit their works. A lot of time is wasted asking about the conference topics, its host country, its ranking, its submission deadline, etc. To tackle this problem, this paper proposes a recommendation approach that suggests personalized upcoming academic venues to computer scientists that fit their current research area and also their interests in terms of venue location and ranking. The target researcher and his community current preferences are taken into consideration. Experiments demonstrate the effectiveness of our proposed rating and recommendation method and show that it outperforms the baseline venue recommendations in terms of accuracy and ranking quality.

Research paper thumbnail of Big Data Classification Using Belief Decision Trees: Application to Intrusion Detection

Over the past few years, the data volume explosion fueled by exciting progression in computer tec... more Over the past few years, the data volume explosion fueled by exciting progression in computer technologies, made the Big Data the focus of widespread attention. Big Data is nebulous since it is an interaction result of several dimensions of scale, among them the veracity which refers to biases and noise in data. Therefore, Big Data veracity is a challenge because it requires a different approach in order to cope with this imperfection. We propose to involve the belief function theory and the belief decison tree as a classification technique to accommodate large applications where the uncertainty reigns. In this paper, we will be firstly concerned with the construction of the belief decision tree, using MapReduce programming model and the averaging approach as a classification method under uncertainty. Then, we will conduct experiments on intrusion detection massive data set, to distinguish between attacks and normal connections in such uncertain context.

Research paper thumbnail of A Gold Standards-Based Crowd Label Aggregation Within the Belief Function Theory

Advances in Artificial Intelligence: From Theory to Practice, 2017

Crowdsourcing, in particular microtasking is now a powerful concept used by employers in order to... more Crowdsourcing, in particular microtasking is now a powerful concept used by employers in order to obtain answers on tasks hardly handled by automated computation. These answers are provided by human employees and then combined to get a final answer. Nevertheless, the quality of participants in microtasking platforms is often heterogeneous which makes results imperfect and thus not fully reliable. To tackle this problem, we propose a new approach of label aggregation based on gold standards under the belief function theory. This latter provides several tools able to represent and even combine imperfect information. Experiments conducted on both simulated and real world datasets show that our approach improves results quality even with a high ratio of bad workers.

Research paper thumbnail of Prédiction des liens dans les réseaux sociaux dans le cadre de la théorie des fonctions de croyance

The link prediction problem is an important research area handled in social network analysis. It ... more The link prediction problem is an important research area handled in social network analysis. It consists of inferring the potential links to be formed in the futur given a current snapshot of the network. Several methods have been proposed to address this problem but most of them consider it under a certain framework. Yet, social networks data are often incomplete and noisy, thus it is necessary to manage uncertainty in the prediction task. We review in this paper, the link prediction problem under uncertainty using the belief function theory. First, we present a new graph based model for social networks that encapsulates the uncertainties in the link structure. Then, we propose a new approach for link prediction through information fusion of the neighboring nodes.

Research paper thumbnail of Towards a Hybrid User and Item-Based Collaborative Filtering Under the Belief Function Theory

Collaborative Filtering (CF) approaches enjoy considerable popularity in the field of Recommender... more Collaborative Filtering (CF) approaches enjoy considerable popularity in the field of Recommender Systems (RSs). They exploit the users’ past ratings and provide personalized recommendations on this basis. Commonly, neighborhood-based CF approaches focus on relationships between items (item-based) or, alternatively, between users (user-based). User-based CF predicts new preferences based on the users sharing similar interests. Item-based computes the similarity between items rather than users to perform the final predictions. However, in both approaches, only partial information from the rating matrix is exploited since they rely either on the ratings of similar users or similar items. Besides, the reliability of the information provided by these pieces of evidence as well as the final predictions cannot be fully trusted. To tackle these issues, we propose a new hybrid neighborhood-based CF under the belief function framework. Our approach tends to take advantage of the two kinds of...

Research paper thumbnail of An Evidential Collaborative Filtering Dealing with Sparsity Problem and Data Imperfections

One of the most promising approaches commonly used in Recommender Systems (RSs) is Collaborative ... more One of the most promising approaches commonly used in Recommender Systems (RSs) is Collaborative Filtering (CF). It relies on a matrix of user-item ratings and makes use of past users’ ratings to generate predictions. Nonetheless, a large amount of ratings in the typical user-item matrix may be unavailable. The insufficiency of available rating data is referred to as the sparsity problem, one of the major issues that limit the quality of recommendations and the applicability of CF. Generally, the final predictions are represented as a certain rating score. This does not reflect the reality which is related to uncertainty and imprecision by nature. Dealing with data imperfections is another fundamental challenge in RSs allowing more reliable and intelligible predictions. Thereupon, we propose in this paper a Collaborative Filtering system that not only tackles the sparsity problem but also deals with data imperfections using the belief function theory.

Research paper thumbnail of A GAN-BERT based decision making approach in peer review

Social network analysis and mining, May 22, 2024

Research paper thumbnail of GAN Based Restyling of Arabic Handwritten Historical Documents

Research paper thumbnail of Towards more trustworthy predictions: A hybrid evidential movie recommender system

JUCS - Journal of Universal Computer Science

Recommender Systems (RSs) are considered as popular tools that have revolutionized the e-commerce... more Recommender Systems (RSs) are considered as popular tools that have revolutionized the e-commerce and digital marketing. Their main goal is predicting the users’ future preferences and providing accessible and personalized recommendations. However, uncertainty can spread at any level throughout the recommendation process, which may affect the results. In fact, the ratings given by the users are often unreliable. The final provided predictions itself may also be pervaded with uncertainty and doubt. Obviously, the reliability of the predictions cannot be fully certain and trustworthy. For the system to be effective, recommendations must inspire trust in the system and provide reliable and credible recommendations. The user may speculate about the uncertainty pervaded behind the given recommendation. He could tend to a reliable recommendation offering him a global overview about his preferences rather than an inappropriate one that contradicts his activities and objectives. While such ...

Research paper thumbnail of A new evidential collaborative filtering: A hybrid memory- and model-based approach

Data Science and Knowledge Engineering for Sensing Decision Support, 2018

Research paper thumbnail of A worker clustering-based approach of label aggregation under the belief function theory

Applied Intelligence, 2018

Crowdsourcing platforms have been attracting a wide attention in the field of artificial intellig... more Crowdsourcing platforms have been attracting a wide attention in the field of artificial intelligence in recent years, providing a cheap and reachable human-powered resource to gather massive labeled data. These data are used to effectively build supervised learning models for academic research puposes. However, despite the attractiveness of these systems, the major concern has always been the quality of the collected labels. In fact, a wide range of workers contributes in labeling data leading to be in possession of potentially noisy and imperfect labels. Therefore in this paper, we propose a new label aggregation technique that allows to determine workers qualities via a clustering process and then represent and combine their labels to estimate the final one under the belief function theory. This latter is notorious for its strength and flexibility when dealing with imperfect information. Experimental results demonstrate that our proposed method outperforms the related work baseline and improves results quality.

Research paper thumbnail of Evidential Item-Based Collaborative Filtering

Lecture Notes in Computer Science, 2016

Recommender Systems (RSs) in particular the collaborative filtering approaches have reached a hig... more Recommender Systems (RSs) in particular the collaborative filtering approaches have reached a high level of popularity. These approaches are designed for predicting the user’s future interests towards unrated items. However, the provided predictions should be taken with restraint because of the uncertainty pervading the real-world problems. Indeed, to not give consideration to such uncertainty may lead to unrepresentative results which can deeply affect the predictions’ accuracy as well as the user’s confidence towards the RS. In order to tackle this issue, we propose in this paper a new evidential item-based collaborative filtering approach. In our approach, we involve the belief function theory tools as well as the Evidential K-Nearest Neighbors (EKNN) classifier to deal with the uncertain aspect of items’ recommendation ignored by the classical methods. The performance of our new recommendation approach is proved through a comparative evaluation with several traditional collaborative filtering recommenders.

Research paper thumbnail of An Evidential Semi-supervised Label Aggregation Approach

Crowdsourcing is a powerful concept that typically takes advantage of human intelligence to deal ... more Crowdsourcing is a powerful concept that typically takes advantage of human intelligence to deal with problems in many fields most importantly in machine learning. Indeed, it enables to collect training labels in a fast and cheap way for supervised algorithms. The only major challenge is that the quality of the contributions is not always guaranteed because of the expertise heterogeneity of the participants. One of the basic strategies to overcome this problem is to assign each task to multiple workers and then combine their answers in order to obtain a single reliable one. This paper provides a new iterative approach that aggregates imperfect labels using the supervision of few gold labels under the evidence theory. Besides of inferring the consensus answers, the workers’ accuracies and the questions difficulties are as well estimated. A comparative evaluation on synthetic and real datasets confirms the effectiveness of our semi-supervised approach over the baselines.

Research paper thumbnail of An Adaptive Approach of Label Aggregation Using a Belief Function Framework

Lecture Notes in Business Information Processing, 2017

Crowdsourcing knows a large expansion in recent years. It is widely used as a low-cost alternativ... more Crowdsourcing knows a large expansion in recent years. It is widely used as a low-cost alternative to guess the true labels of training data in machine learning problems. In fact, crowdsourcing platforms such as Amazon’s Mechanical Turk allow to collect from crowd workers multiple labels aggregated thereafter to infer the true label. As the workers are not always reliable, imperfect labels can occur. In this work, we propose an approach that aggregates labels using the belief function theory besides of adaptively integrating both labelers expertise and question difficulty. Experiments with real data demonstrate that our method provides better aggregation results.

Research paper thumbnail of Academic Venue Recommendation Based on Refined Cross Domain

Intelligent Systems Design and Applications, 2022

Research paper thumbnail of Big Data Classification Using Belief Decision Trees: Application to Intrusion Detection

Advances in Intelligent Systems and Computing, 2015

Over the past few years, the data volume explosion fueled by exciting progression in computer tec... more Over the past few years, the data volume explosion fueled by exciting progression in computer technologies, made the Big Data the focus of widespread attention. Big Data is nebulous since it is an interaction result of several dimensions of scale, among them the veracity which refers to biases and noise in data. Therefore, Big Data veracity is a challenge because it requires a different approach in order to cope with this imperfection. We propose to involve the belief function theory and the belief decison tree as a classification technique to accommodate large applications where the uncertainty reigns. In this paper, we will be firstly concerned with the construction of the belief decision tree, using MapReduce programming model and the averaging approach as a classification method under uncertainty. Then, we will conduct experiments on intrusion detection massive data set, to distinguish between attacks and normal connections in such uncertain context.

Research paper thumbnail of A New User-Based Collaborative Filtering Under the Belief Function Theory

Advances in Artificial Intelligence: From Theory to Practice, 2017

The collaborative filtering (CF) is considered as the most widely used approach in the field of R... more The collaborative filtering (CF) is considered as the most widely used approach in the field of Recommender Systems (RSs). It tends to predict the users’ preferences based on the users sharing similar interests. However, ignoring the uncertainty involved in the provided predictions is among the limitations related to this approach. To deal with this issue, we propose in this paper a new user-based collaborative filtering within the belief function theory. In our approach, the evidence of each similar user is taken into account and Dempster’s rule of combination is used for combining these pieces of evidence. A comparative evaluation on a real world data set shows that the proposed method outperforms traditional user-based collaborative filtering recommenders.

Research paper thumbnail of Iterative Aggregation of Crowdsourced Tasks Within the Belief Function Theory

Lecture Notes in Computer Science, 2017

With the growing of crowdsourcing services, gathering training data for supervised machine learni... more With the growing of crowdsourcing services, gathering training data for supervised machine learning has become cheaper and faster than engaging experts. However, the quality of the crowd-generated labels remains an open issue. This is basically due to the wide ranging expertise levels of the participants in the labeling process. In this paper, we present an iterative approach of label aggregation based on the belief function theory that simultanously estimates labels, the reliability of participants and difficulty of each task. Our empirical evaluation demonstrate the efficiency of our method as it gives better quality labels.

Research paper thumbnail of An Evidential Imprecise Answer Aggregation Approach Based on Worker Clustering

Intelligent Data Engineering and Automated Learning – IDEAL 2019, 2019

Crowdsourcing has become a popular and practical tool to gather low-cost labels from human worker... more Crowdsourcing has become a popular and practical tool to gather low-cost labels from human workers in order to provide training data for machine learning applications. However, the quality of the crowdsourced data has always been an issue mainly caused by the quality of the contributors. Since they can be unreliable due to many factors, it became common to assign a task to more than one person and then combine the gathered contributions in order to obtain high quality results. In this work, we propose a new approach of answer combination within an evidential framework to cope with uncertainty. In fact, we assume that answers could be partial which means imprecise or even incomplete. Moreover, the approach includes an important step that clusters workers using the k-means algorithm to determine their types in order to effectively integrate them in the aggregation of answers step. Experimentation on simulated dataset show the efficiency of our approach to improve outcome quality.

Research paper thumbnail of An Evidential Collaborative Filtering Approach Based on Items Contents Clustering

Belief Functions: Theory and Applications, 2018

Research paper thumbnail of Exploring Location and Ranking for Academic Venue Recommendation

Advances in Intelligent Systems and Computing, 2018

Publishing scientific results is extremely important for each researcher. The concrete challenge ... more Publishing scientific results is extremely important for each researcher. The concrete challenge is how to select the right academic venue that corresponds to researcher's current interest and without missing the deadline at the same time. Due to the huge number of academic venues especially in the field of computer science, it is difficult for researchers to choose a conference or a journal to submit their works. A lot of time is wasted asking about the conference topics, its host country, its ranking, its submission deadline, etc. To tackle this problem, this paper proposes a recommendation approach that suggests personalized upcoming academic venues to computer scientists that fit their current research area and also their interests in terms of venue location and ranking. The target researcher and his community current preferences are taken into consideration. Experiments demonstrate the effectiveness of our proposed rating and recommendation method and show that it outperforms the baseline venue recommendations in terms of accuracy and ranking quality.

Research paper thumbnail of Big Data Classification Using Belief Decision Trees: Application to Intrusion Detection

Over the past few years, the data volume explosion fueled by exciting progression in computer tec... more Over the past few years, the data volume explosion fueled by exciting progression in computer technologies, made the Big Data the focus of widespread attention. Big Data is nebulous since it is an interaction result of several dimensions of scale, among them the veracity which refers to biases and noise in data. Therefore, Big Data veracity is a challenge because it requires a different approach in order to cope with this imperfection. We propose to involve the belief function theory and the belief decison tree as a classification technique to accommodate large applications where the uncertainty reigns. In this paper, we will be firstly concerned with the construction of the belief decision tree, using MapReduce programming model and the averaging approach as a classification method under uncertainty. Then, we will conduct experiments on intrusion detection massive data set, to distinguish between attacks and normal connections in such uncertain context.

Research paper thumbnail of A Gold Standards-Based Crowd Label Aggregation Within the Belief Function Theory

Advances in Artificial Intelligence: From Theory to Practice, 2017

Crowdsourcing, in particular microtasking is now a powerful concept used by employers in order to... more Crowdsourcing, in particular microtasking is now a powerful concept used by employers in order to obtain answers on tasks hardly handled by automated computation. These answers are provided by human employees and then combined to get a final answer. Nevertheless, the quality of participants in microtasking platforms is often heterogeneous which makes results imperfect and thus not fully reliable. To tackle this problem, we propose a new approach of label aggregation based on gold standards under the belief function theory. This latter provides several tools able to represent and even combine imperfect information. Experiments conducted on both simulated and real world datasets show that our approach improves results quality even with a high ratio of bad workers.

Research paper thumbnail of Prédiction des liens dans les réseaux sociaux dans le cadre de la théorie des fonctions de croyance

The link prediction problem is an important research area handled in social network analysis. It ... more The link prediction problem is an important research area handled in social network analysis. It consists of inferring the potential links to be formed in the futur given a current snapshot of the network. Several methods have been proposed to address this problem but most of them consider it under a certain framework. Yet, social networks data are often incomplete and noisy, thus it is necessary to manage uncertainty in the prediction task. We review in this paper, the link prediction problem under uncertainty using the belief function theory. First, we present a new graph based model for social networks that encapsulates the uncertainties in the link structure. Then, we propose a new approach for link prediction through information fusion of the neighboring nodes.

Research paper thumbnail of Towards a Hybrid User and Item-Based Collaborative Filtering Under the Belief Function Theory

Collaborative Filtering (CF) approaches enjoy considerable popularity in the field of Recommender... more Collaborative Filtering (CF) approaches enjoy considerable popularity in the field of Recommender Systems (RSs). They exploit the users’ past ratings and provide personalized recommendations on this basis. Commonly, neighborhood-based CF approaches focus on relationships between items (item-based) or, alternatively, between users (user-based). User-based CF predicts new preferences based on the users sharing similar interests. Item-based computes the similarity between items rather than users to perform the final predictions. However, in both approaches, only partial information from the rating matrix is exploited since they rely either on the ratings of similar users or similar items. Besides, the reliability of the information provided by these pieces of evidence as well as the final predictions cannot be fully trusted. To tackle these issues, we propose a new hybrid neighborhood-based CF under the belief function framework. Our approach tends to take advantage of the two kinds of...

Research paper thumbnail of An Evidential Collaborative Filtering Dealing with Sparsity Problem and Data Imperfections

One of the most promising approaches commonly used in Recommender Systems (RSs) is Collaborative ... more One of the most promising approaches commonly used in Recommender Systems (RSs) is Collaborative Filtering (CF). It relies on a matrix of user-item ratings and makes use of past users’ ratings to generate predictions. Nonetheless, a large amount of ratings in the typical user-item matrix may be unavailable. The insufficiency of available rating data is referred to as the sparsity problem, one of the major issues that limit the quality of recommendations and the applicability of CF. Generally, the final predictions are represented as a certain rating score. This does not reflect the reality which is related to uncertainty and imprecision by nature. Dealing with data imperfections is another fundamental challenge in RSs allowing more reliable and intelligible predictions. Thereupon, we propose in this paper a Collaborative Filtering system that not only tackles the sparsity problem but also deals with data imperfections using the belief function theory.