Concurrency and Computation: Practice and Experience, 2022
Multi‐criteria recommender systems (MCRSs) provide suggestions to users based on their preference... more Multi‐criteria recommender systems (MCRSs) provide suggestions to users based on their preferences to various criteria. Incorporation of criteria ratings into recommendation framework can provide quality recommendations to users because these ratings can elicit users' preferences efficiently. However, elicitation of user's overall preference based on criteria ratings is a key issue in MCRS. Even though several aggregation methods for the elicitation of users' overall preference have been investigated in the literature, no method has been shown the superiority under all circumstances. Therefore, we propose a model based approach to user preference discovery in multi‐criteria RS using genetic programming (GP). In this work, we suggest three‐stage process to generate recommendations to users. First, we learn user preference transformation function to aggregate criteria ratings by using GP, and then we utilize the preference function, so derived, for computing similarities in MCRS. Finally, items are recommended to users. Experimental results on Yahoo! Movies dataset show the superiority of our proposed approach in comparison to other aggregation approaches.
2020 5th International Conference on Communication and Electronics Systems (ICCES), 2020
Recommender systems (RSs) are software tools that work as guides by suggesting products to users ... more Recommender systems (RSs) are software tools that work as guides by suggesting products to users from a vast catalogue of products. Various approaches and techniques have been developed to provide effective recommendations to users. Classical collaborative filtering (CF) based RSs helps users by providing suggestions based on their overall assessment of items. However, providing suggestions based on their overall assessment is not an efficient way. So, multi-criteria recommender systems (MCRS) came into existence as an extended approach for suggesting products to users based on multiple features of products, and adding these multiple features can enhance the performance of the system. However, aggregation of these feature assessment i.e. feedback provided to multiple criteria is a key issue in MCRS. In this paper, we present a comparative analysis of genetic algorithm (GA) and genetic programming (GP) approaches to aggregate criteria ratings for predicting user preferences in MCRS. These two algorithms are bio-inspired and have great potential to solve optimization problems. In this research, GP and GA are used to solve the aggregation problem in MCRS by estimating weights for each criterion in a system. We compared the results of genetic programming and genetic algorithm approaches to show their effectiveness in multi-criteria rating systems.
Information and Communication Technology for Intelligent Systems (ICTIS 2017) - Volume 1, 2017
Recommender systems (RS) try to solve information overload problem by providing the most relevant... more Recommender systems (RS) try to solve information overload problem by providing the most relevant items to users from a large set of items. Collaborative filtering (CF), a popular approach in building RS, generates recommendations to users based on explicit ratings provided by the community of users. Currently many online platforms allow users to evaluate items based on multiple criteria along with an overall rating instead of single overall rating. Previous research work has shown that considering these multiple criteria ratings for recommendations improved the predictive accuracy of recommender systems.
Advances in Intelligent Systems and Computing, 2017
Recommender system aims to solve the information overload problem by recommending a set of items ... more Recommender system aims to solve the information overload problem by recommending a set of items that are suitable for users. Recently, the incorporation of multiple criteria into traditional single-criterion recommender system has increased the interest. In this paper, we propose a novel trust-enhanced multi-criteria recommender system using fuzzy rating in collaborative filtering framework. We have also designed a hybrid approach of traditional multi-criteria recommender system and trust-enhanced multi-criteria recommender system to reduce data sparsity problem. The empirical results show that our proposed approach demonstrates efficient recommendation as compared to traditional approach.
Mining Intelligence and Knowledge Exploration, 2017
Collaborative filtering (CF), the most prevalent technique in the area of recommender systems (RS... more Collaborative filtering (CF), the most prevalent technique in the area of recommender systems (RSs), provides suggestions to users based on the tastes of their similar users. However, the new user and sparsity problems, degrade its efficiency of recommendations. Trust can enhance the recommendation quality by mimicking social dictum "friend of a friend will be a friend". However distrust, the another face of coin is yet to be explored along with trust in the area of RSs. Our work in this paper is an attempt toward introducing trust-distrust enhanced recommendations based on the novel similarity measure that combines user ratings and trust values for generating more quality recommendations. Our approach also exploits distrust links among users and analyses their propagation effects. Further, distrust values are also used for filtering more distrust-worthy neighbours from the neighbourhood set. Our experimental results show that our proposed approaches outperform the traditional CF and existing trust enhanced approaches in terms of various performance measures.
2020 4th International Conference on Intelligent Computing and Control Systems (ICICCS), 2020
Recommender systems (RSs) are personalization tools that gives recommendations for items to users... more Recommender systems (RSs) are personalization tools that gives recommendations for items to users by exploiting various methods. Conventional collaborative filtering (CF) based RSs provide suggestions to users based on overall rating of items which is not an efficient procedure as users in system may have different choices on different criteria. So, multicriteria recommender systems (MCRS) came into existence as an extension of traditional CF based RSs. MCRS recommends items to users based on number of criteria. Recommending products to users from the vast catalog is still a challenge for researchers. This paper presents a review of some significant work in the area of multi-criteria recommender system. After a brief introduction, we present review of existing methods categories according to heuristic and model based approach, and some of the popular approaches are classified into different sets such as recommendation fields, research problem, data mining and machine learning techniques. Insights and possible future work in the area of MCRSs are also discussed.
Recommendation systems are the agents that help the learner to identify a subset of suitable lear... more Recommendation systems are the agents that help the learner to identify a subset of suitable learning resources from a variety of choices. Recommendation Systems is a widely explored field since the last decade. Much of the work is going on in recommendation systems that are based on the evaluation of resources and users‟ data. In this paper we concentrate on E-Learning Recommendation Systems. An E-learning recommendation system is a derivative field of recommendation systems in which the resources are specifically the available bulk of learning material either online or offline. The aim of E-learning software is to select the useful piece of material which the learner actually requires to study. Our aim in this paper is to study various recommendation techniques with their virtues and shortcomings. Further we will discuss E-learning recommendation systems with a brief review of some major milestones in the field of E-Learning.
Communications in Computer and Information Science, 2019
Cyberbullying is one of the radical emerging problems with the advancements in the Internet, conn... more Cyberbullying is one of the radical emerging problems with the advancements in the Internet, connecting people around the globe by social media networks. Existing studies mostly focus only on cyberbullying detection in the English language, thus the main objective of this paper is to develop an approach to detect cyberbullying in Hindi-English code-mixed language (Hinglish) which is exorbitantly used by Indian users. Due to the unavailability of Hinglish dataset, we created the Hinglish Cyberbullying Comments (HCC) labeled dataset consisting of comments from social media networks such as Instagram and YouTube. We also developed eight different machine learning models for sentiment classification in-order to automatically detect incidents of cyberbullying. Performance measures namely accuracy, precision, recall and f1 score are used to evaluate these models. Eventually, a hybrid model is developed based on top performers of these eight baseline classifiers which perform better with an accuracy of 80.26% and f1-score of 82.96%.
Group recommender systems (GRSs) have played an important role in numerous online applications by... more Group recommender systems (GRSs) have played an important role in numerous online applications by providing recommendation to the group of users where satisfaction of the entire group is a major concern. In traditional GRSs, the relevance of all the groups and the items is considered equal which does not produce accurate recommendations. In this paper, we propose a formalization of the GRS based on the relevance concept using profile merging scheme where collaborative filtering (CF) is applied on each group profile to generate effective recommendations to the group by considering the ratings of the items, the relevance of the groups and the relevance of the items. Further, our GRS framework provides relevant similarity measures, relevant prediction and recommendation quality measures. The experimental results on the benchmark MovieLens dataset demonstrate the efficacy of our proposed GRS framework.
2017 International Conference on Advances in Computing, Communications and Informatics (ICACCI), 2017
Emotion recognition system using electrocardiogram (ECG) has received considerable attention rece... more Emotion recognition system using electrocardiogram (ECG) has received considerable attention recently in the area of human computer interaction (HCI). Our work in this paper is an attempt towards developing an emotion recognition system that would classify emotions effectively into four emotional states: joy, anger, sadness and pleasure. The contributions of this paper is summarized in three fold: Firstly, we extract statistical features through digital filter. Secondly, to extract spectral features such as power and entropy features, we decompose and reconstruct the ECG signal through empirical mode decomposition (EMD) and apply Hilbert huang transform (HHT) as well as discrete fourier transform (DFT) to the intrinsic mode functions (IMFs). Finally, the effectiveness of our proposed hybrid scheme is demonstrated through experimental results in terms of various performance measures.
Collaborative filtering (CF) has been investigated and improved extensively over the past years b... more Collaborative filtering (CF) has been investigated and improved extensively over the past years but still unable to handle multiple issues like cold-start and sparsity problems due to the absence of user-item rating information. Further, it has been seen that the contextual information plays a significant role for generating user relevant situational recommendations but the incorporation of contextual information into CF directly is the major problem in RS. This paper is an effort toward developing recommendation strategy based on contextual fuzzy CF by utilising particle swarm optimisation (PSO) algorithm. This work has been completed in two-fold. First, we incorporate contextual information into fuzzy CF algorithm through context modelling approach. Second, we extend the previous method by employing PSO algorithm in order to learn user weights on various hybrid fuzzy features for enhancing the performance of CF technique. The results show the superiority of our proposed method against other comparative methods.
Proceedings of the 9th International Conference on Machine Learning and Computing, 2017
Recommender system (RS), a web personalization tool, attempts to generate suitable recommendation... more Recommender system (RS), a web personalization tool, attempts to generate suitable recommendations to users based on their preferences. Generally, recommender system works on overall ratings but these ratings do not reflect the actual user preferences. Therefore, incorporation of multiple criteria ratings into RS can capture the user preferences accurately and produce effective recommendations to users. Multi criteria recommender systems (MCRS) generate recommendations to users based on the aggregation of similarities computed on multiple criteria using collaborative filtering. However, capturing optimal weights of various users on different criteria in the process of similarity aggregation is a major concern. Further selection of appropriate similarity measure is another challenge for employing collaborative filtering. Our work in this paper is an attempt towards developing multi criteria recommender systems by utilizing various similarity measures and particle swarm optimization to learn optimal weights. Experimental results reveal that our proposed approaches outperform other traditional approaches.
Multi criteria recommender systems generate quality recommendations to users by incorporating cri... more Multi criteria recommender systems generate quality recommendations to users by incorporating criteria ratings into recommender system using collaborative filtering because ratings over multiple criteria can capture user preferences efficiently. However, aggregation of similarities computed on multiple criteria is still a major concern. Moreover, the concept of natural noise is an emerging trend that is related to inconsistent behaviour of users. Our work in this paper is an attempt towards developing multi criteria recommender systems that deals with inconsistent ratings and uses particle swarm optimization to learn optimal weights for a user over different criteria in the aggregation process.
Recommender system has been emerged as a personalization tool to solve the issue of information o... more Recommender system has been emerged as a personalization tool to solve the issue of information overload in an e-commerce environment. Traditional collaborative filtering (CF) based recommender systems (RSs) suggest items to users based on their overall ratings which are used to find out similar users. Multi-criteria ratings are used to capture user preferences efficiently in multi-criteria recommender systems (MCRS), and incorporation of various criteria ratings can lead to higher performance in MCRS. Usually, user relies on the credibility of an item provided through his/her social circle or similar users, which is called a personal view on items from their close ones. However, it is not generally sufficient to depend exclusively on the personal view of the user. Therefore, public view that includes whole community can play a key role in the credibility of an item. In this paper, we propose a MCRS based on the credibility score of an item, which is an aggregated value of credibility scores on various criteria of an item. These credibility scores are computed based on personal and public views. However, different users have different priorities to various criteria of an item. Therefore, we use genetic algorithm (GA) to learn appropriate weights in the aggregation task of credibility score. The experiment results on Yahoo! Movies and modified MovieLens dataset demonstrate the effectiveness of proposed credibility score based MCRS in terms of coverage, recall, precision, and f-measure.
Recommender system is one of the emerging personalization tools in e-commerce domains for suggest... more Recommender system is one of the emerging personalization tools in e-commerce domains for suggesting suitable items to users. Traditional collaborative filtering (CF) based recommender systems (RSs) suggest items to users based on the overall ratings to find out similar users. Multi-criteria ratings are used to capture user preferences efficiently in multi-criteria recommender systems (MCRSs), and incorporation of criteria ratings can lead to higher performance in MCRS. However, aggregation of these criteria ratings is a major concern in MCRS. In this paper, we propose a multi-criteria collaborative filtering-based RS by leveraging information derived from multi-criteria ratings through Genetic programming (GP). The proposed system consists of two parts: (1) weights of each user for every criterion are computed through our proposed modified sub-tree crossover in GP process (2) criteria weights are then incorporated in CF process to generate effective recommendations in our proposed system. The obtained results present significant improvements in prediction and recommendation qualities in comparison to heuristic approaches.
Journal of Statistics and Management Systems, 2019
Facial Gender Identification has vast application in human computer interaction, determines custo... more Facial Gender Identification has vast application in human computer interaction, determines customer profile in shopping centers, and restricted permission to enter in prohibited zone, criminal profile analysis. This paper presents a robust process for illumination invariant compact feature extraction using Gabor filter for the automatic
Load forecasting is a challenging task in power markets that require attention in generating accu... more Load forecasting is a challenging task in power markets that require attention in generating accurate and stable load to deal with planning and management strategies. In past few years, several intelligence-based models have been introduced for precise load forecast. Among them, artificial neural network (ANN) seems more effective and capable to handle the non-linear behavior of load and generates an accurate forecast. However, it suffers from overfitting problem thus reducing the accuracy of load forecasts. To overcome this problem, a hybrid methodology namely ANN-IEAMCGM-R, for short-term load forecast is proposed in this paper. ANN is integrated with an enhanced evolutionary algorithm (IEAMCGM-R) to find optimal network weights. This evolutionary algorithm is composed of improved environmental adaptation method with real parameters (IEAM-R) and our proposed Controlled Gaussian Mutation (CGM) method to bring greater diversity within the population resulting in a higher convergence of solutions. The electric load data from the New England Power Pool (NEPOOL, ISO New England) and Australian Energy Market Operator (New South Wales (NSW), Australia) have been used to illustrate the efficacy of the proposed hybrid methodology. Results show that the proposed hybrid methodology generates higher accuracy than other state-of-theart algorithms.
Collaborative filleting is one of the widely implemented techniques in the area of recommender sy... more Collaborative filleting is one of the widely implemented techniques in the area of recommender systems. But it suffers from data sparsity problem. To address that problem, cross-domain recommender systems (CDRSs) have been emerged to solve the data sparsity problem and improve the accuracy of prediction by transfer learning mechanism. To apply transfer learning mechanism, some common properties associated with users and/or items are needed between the domains. Several attempts have shown that recommendation quality of cross-domain recommender systems could be improved by transferring the user-generated tag information into the target domain. However, sometimes that information is not enough to accomplish recommendation task efficiently. To this end, item features can also be a valuable source of information for developing the correlation between domains and would be considered in generating effective recommendations in target domain. In this paper, we propose a model by utilizing item features and user-generated tags through matrix factorization in CDRSs framework. Firstly, we extract item features in terms of genres and user preferences in terms of user-generated tags. Thereafter, to establish the bridge for transferring knowledge, matrix factorization has been used. Finally, experimental results demonstrate that our proposed model outperforms the other single domain as well as cross domain approaches in CDRSs framework.
In the area of recommender systems, collaborative filtering is widely used technique for recommen... more In the area of recommender systems, collaborative filtering is widely used technique for recommending appropriate items to a user based on the available ratings given by similar users. Most recommender systems (RSs) work only on the single criterion rating i.e., overall rating, however overall rating may not be a good representative of a user preference. Single criterion collaborative filtering (CF) does not generate more reliable recommendations because it suffers from correlation based similarity problems. Moreover, representation of uncertain user preferences is another concern of CF. In our work, we develop a novel fuzzy Bayesian approach to multi-criteria CF for handling uncertain user preferences and correlation based similarity problems. Further, incorporation of multicriteria ratings into CF would be helpful for generating effective recommendations. Through experiments on Yahoo! Movies dataset, we compare our proposed approach to baseline approaches and demonstrate its effectiveness in terms of accuracy, recall and f-measure.
With the rapid advancement of information and communication technologies, e-learning has gained a... more With the rapid advancement of information and communication technologies, e-learning has gained a considerable attention in recent years. Many researchers have attempted to develop various e-learning systems with personalized learning mechanisms for assisting learners so that they can learn more efficiently. In this context, curriculum sequencing is considered as an important concern for developing more efficient personalized e-learning systems. A more effective personalized e-learning recommender system should recommend a sequence of learning materials called learning path, in an appropriate order with a starting and ending point, rather than a sequence of unordered learning materials. Further the recommended sequence should also match the learner preferences for enhancing their learning capabilities. Moreover, the length of recommended sequence cannot be fixed for each learner because these learners differ from one another in their preferences such as knowledge levels, learning styles, emotions, etc. In this paper, we present an effective learning path recommendation system (LPRS) for e-learners through a variable length genetic algorithm (VLGA) by considering learners'
Concurrency and Computation: Practice and Experience, 2022
Multi‐criteria recommender systems (MCRSs) provide suggestions to users based on their preference... more Multi‐criteria recommender systems (MCRSs) provide suggestions to users based on their preferences to various criteria. Incorporation of criteria ratings into recommendation framework can provide quality recommendations to users because these ratings can elicit users' preferences efficiently. However, elicitation of user's overall preference based on criteria ratings is a key issue in MCRS. Even though several aggregation methods for the elicitation of users' overall preference have been investigated in the literature, no method has been shown the superiority under all circumstances. Therefore, we propose a model based approach to user preference discovery in multi‐criteria RS using genetic programming (GP). In this work, we suggest three‐stage process to generate recommendations to users. First, we learn user preference transformation function to aggregate criteria ratings by using GP, and then we utilize the preference function, so derived, for computing similarities in MCRS. Finally, items are recommended to users. Experimental results on Yahoo! Movies dataset show the superiority of our proposed approach in comparison to other aggregation approaches.
2020 5th International Conference on Communication and Electronics Systems (ICCES), 2020
Recommender systems (RSs) are software tools that work as guides by suggesting products to users ... more Recommender systems (RSs) are software tools that work as guides by suggesting products to users from a vast catalogue of products. Various approaches and techniques have been developed to provide effective recommendations to users. Classical collaborative filtering (CF) based RSs helps users by providing suggestions based on their overall assessment of items. However, providing suggestions based on their overall assessment is not an efficient way. So, multi-criteria recommender systems (MCRS) came into existence as an extended approach for suggesting products to users based on multiple features of products, and adding these multiple features can enhance the performance of the system. However, aggregation of these feature assessment i.e. feedback provided to multiple criteria is a key issue in MCRS. In this paper, we present a comparative analysis of genetic algorithm (GA) and genetic programming (GP) approaches to aggregate criteria ratings for predicting user preferences in MCRS. These two algorithms are bio-inspired and have great potential to solve optimization problems. In this research, GP and GA are used to solve the aggregation problem in MCRS by estimating weights for each criterion in a system. We compared the results of genetic programming and genetic algorithm approaches to show their effectiveness in multi-criteria rating systems.
Information and Communication Technology for Intelligent Systems (ICTIS 2017) - Volume 1, 2017
Recommender systems (RS) try to solve information overload problem by providing the most relevant... more Recommender systems (RS) try to solve information overload problem by providing the most relevant items to users from a large set of items. Collaborative filtering (CF), a popular approach in building RS, generates recommendations to users based on explicit ratings provided by the community of users. Currently many online platforms allow users to evaluate items based on multiple criteria along with an overall rating instead of single overall rating. Previous research work has shown that considering these multiple criteria ratings for recommendations improved the predictive accuracy of recommender systems.
Advances in Intelligent Systems and Computing, 2017
Recommender system aims to solve the information overload problem by recommending a set of items ... more Recommender system aims to solve the information overload problem by recommending a set of items that are suitable for users. Recently, the incorporation of multiple criteria into traditional single-criterion recommender system has increased the interest. In this paper, we propose a novel trust-enhanced multi-criteria recommender system using fuzzy rating in collaborative filtering framework. We have also designed a hybrid approach of traditional multi-criteria recommender system and trust-enhanced multi-criteria recommender system to reduce data sparsity problem. The empirical results show that our proposed approach demonstrates efficient recommendation as compared to traditional approach.
Mining Intelligence and Knowledge Exploration, 2017
Collaborative filtering (CF), the most prevalent technique in the area of recommender systems (RS... more Collaborative filtering (CF), the most prevalent technique in the area of recommender systems (RSs), provides suggestions to users based on the tastes of their similar users. However, the new user and sparsity problems, degrade its efficiency of recommendations. Trust can enhance the recommendation quality by mimicking social dictum "friend of a friend will be a friend". However distrust, the another face of coin is yet to be explored along with trust in the area of RSs. Our work in this paper is an attempt toward introducing trust-distrust enhanced recommendations based on the novel similarity measure that combines user ratings and trust values for generating more quality recommendations. Our approach also exploits distrust links among users and analyses their propagation effects. Further, distrust values are also used for filtering more distrust-worthy neighbours from the neighbourhood set. Our experimental results show that our proposed approaches outperform the traditional CF and existing trust enhanced approaches in terms of various performance measures.
2020 4th International Conference on Intelligent Computing and Control Systems (ICICCS), 2020
Recommender systems (RSs) are personalization tools that gives recommendations for items to users... more Recommender systems (RSs) are personalization tools that gives recommendations for items to users by exploiting various methods. Conventional collaborative filtering (CF) based RSs provide suggestions to users based on overall rating of items which is not an efficient procedure as users in system may have different choices on different criteria. So, multicriteria recommender systems (MCRS) came into existence as an extension of traditional CF based RSs. MCRS recommends items to users based on number of criteria. Recommending products to users from the vast catalog is still a challenge for researchers. This paper presents a review of some significant work in the area of multi-criteria recommender system. After a brief introduction, we present review of existing methods categories according to heuristic and model based approach, and some of the popular approaches are classified into different sets such as recommendation fields, research problem, data mining and machine learning techniques. Insights and possible future work in the area of MCRSs are also discussed.
Recommendation systems are the agents that help the learner to identify a subset of suitable lear... more Recommendation systems are the agents that help the learner to identify a subset of suitable learning resources from a variety of choices. Recommendation Systems is a widely explored field since the last decade. Much of the work is going on in recommendation systems that are based on the evaluation of resources and users‟ data. In this paper we concentrate on E-Learning Recommendation Systems. An E-learning recommendation system is a derivative field of recommendation systems in which the resources are specifically the available bulk of learning material either online or offline. The aim of E-learning software is to select the useful piece of material which the learner actually requires to study. Our aim in this paper is to study various recommendation techniques with their virtues and shortcomings. Further we will discuss E-learning recommendation systems with a brief review of some major milestones in the field of E-Learning.
Communications in Computer and Information Science, 2019
Cyberbullying is one of the radical emerging problems with the advancements in the Internet, conn... more Cyberbullying is one of the radical emerging problems with the advancements in the Internet, connecting people around the globe by social media networks. Existing studies mostly focus only on cyberbullying detection in the English language, thus the main objective of this paper is to develop an approach to detect cyberbullying in Hindi-English code-mixed language (Hinglish) which is exorbitantly used by Indian users. Due to the unavailability of Hinglish dataset, we created the Hinglish Cyberbullying Comments (HCC) labeled dataset consisting of comments from social media networks such as Instagram and YouTube. We also developed eight different machine learning models for sentiment classification in-order to automatically detect incidents of cyberbullying. Performance measures namely accuracy, precision, recall and f1 score are used to evaluate these models. Eventually, a hybrid model is developed based on top performers of these eight baseline classifiers which perform better with an accuracy of 80.26% and f1-score of 82.96%.
Group recommender systems (GRSs) have played an important role in numerous online applications by... more Group recommender systems (GRSs) have played an important role in numerous online applications by providing recommendation to the group of users where satisfaction of the entire group is a major concern. In traditional GRSs, the relevance of all the groups and the items is considered equal which does not produce accurate recommendations. In this paper, we propose a formalization of the GRS based on the relevance concept using profile merging scheme where collaborative filtering (CF) is applied on each group profile to generate effective recommendations to the group by considering the ratings of the items, the relevance of the groups and the relevance of the items. Further, our GRS framework provides relevant similarity measures, relevant prediction and recommendation quality measures. The experimental results on the benchmark MovieLens dataset demonstrate the efficacy of our proposed GRS framework.
2017 International Conference on Advances in Computing, Communications and Informatics (ICACCI), 2017
Emotion recognition system using electrocardiogram (ECG) has received considerable attention rece... more Emotion recognition system using electrocardiogram (ECG) has received considerable attention recently in the area of human computer interaction (HCI). Our work in this paper is an attempt towards developing an emotion recognition system that would classify emotions effectively into four emotional states: joy, anger, sadness and pleasure. The contributions of this paper is summarized in three fold: Firstly, we extract statistical features through digital filter. Secondly, to extract spectral features such as power and entropy features, we decompose and reconstruct the ECG signal through empirical mode decomposition (EMD) and apply Hilbert huang transform (HHT) as well as discrete fourier transform (DFT) to the intrinsic mode functions (IMFs). Finally, the effectiveness of our proposed hybrid scheme is demonstrated through experimental results in terms of various performance measures.
Collaborative filtering (CF) has been investigated and improved extensively over the past years b... more Collaborative filtering (CF) has been investigated and improved extensively over the past years but still unable to handle multiple issues like cold-start and sparsity problems due to the absence of user-item rating information. Further, it has been seen that the contextual information plays a significant role for generating user relevant situational recommendations but the incorporation of contextual information into CF directly is the major problem in RS. This paper is an effort toward developing recommendation strategy based on contextual fuzzy CF by utilising particle swarm optimisation (PSO) algorithm. This work has been completed in two-fold. First, we incorporate contextual information into fuzzy CF algorithm through context modelling approach. Second, we extend the previous method by employing PSO algorithm in order to learn user weights on various hybrid fuzzy features for enhancing the performance of CF technique. The results show the superiority of our proposed method against other comparative methods.
Proceedings of the 9th International Conference on Machine Learning and Computing, 2017
Recommender system (RS), a web personalization tool, attempts to generate suitable recommendation... more Recommender system (RS), a web personalization tool, attempts to generate suitable recommendations to users based on their preferences. Generally, recommender system works on overall ratings but these ratings do not reflect the actual user preferences. Therefore, incorporation of multiple criteria ratings into RS can capture the user preferences accurately and produce effective recommendations to users. Multi criteria recommender systems (MCRS) generate recommendations to users based on the aggregation of similarities computed on multiple criteria using collaborative filtering. However, capturing optimal weights of various users on different criteria in the process of similarity aggregation is a major concern. Further selection of appropriate similarity measure is another challenge for employing collaborative filtering. Our work in this paper is an attempt towards developing multi criteria recommender systems by utilizing various similarity measures and particle swarm optimization to learn optimal weights. Experimental results reveal that our proposed approaches outperform other traditional approaches.
Multi criteria recommender systems generate quality recommendations to users by incorporating cri... more Multi criteria recommender systems generate quality recommendations to users by incorporating criteria ratings into recommender system using collaborative filtering because ratings over multiple criteria can capture user preferences efficiently. However, aggregation of similarities computed on multiple criteria is still a major concern. Moreover, the concept of natural noise is an emerging trend that is related to inconsistent behaviour of users. Our work in this paper is an attempt towards developing multi criteria recommender systems that deals with inconsistent ratings and uses particle swarm optimization to learn optimal weights for a user over different criteria in the aggregation process.
Recommender system has been emerged as a personalization tool to solve the issue of information o... more Recommender system has been emerged as a personalization tool to solve the issue of information overload in an e-commerce environment. Traditional collaborative filtering (CF) based recommender systems (RSs) suggest items to users based on their overall ratings which are used to find out similar users. Multi-criteria ratings are used to capture user preferences efficiently in multi-criteria recommender systems (MCRS), and incorporation of various criteria ratings can lead to higher performance in MCRS. Usually, user relies on the credibility of an item provided through his/her social circle or similar users, which is called a personal view on items from their close ones. However, it is not generally sufficient to depend exclusively on the personal view of the user. Therefore, public view that includes whole community can play a key role in the credibility of an item. In this paper, we propose a MCRS based on the credibility score of an item, which is an aggregated value of credibility scores on various criteria of an item. These credibility scores are computed based on personal and public views. However, different users have different priorities to various criteria of an item. Therefore, we use genetic algorithm (GA) to learn appropriate weights in the aggregation task of credibility score. The experiment results on Yahoo! Movies and modified MovieLens dataset demonstrate the effectiveness of proposed credibility score based MCRS in terms of coverage, recall, precision, and f-measure.
Recommender system is one of the emerging personalization tools in e-commerce domains for suggest... more Recommender system is one of the emerging personalization tools in e-commerce domains for suggesting suitable items to users. Traditional collaborative filtering (CF) based recommender systems (RSs) suggest items to users based on the overall ratings to find out similar users. Multi-criteria ratings are used to capture user preferences efficiently in multi-criteria recommender systems (MCRSs), and incorporation of criteria ratings can lead to higher performance in MCRS. However, aggregation of these criteria ratings is a major concern in MCRS. In this paper, we propose a multi-criteria collaborative filtering-based RS by leveraging information derived from multi-criteria ratings through Genetic programming (GP). The proposed system consists of two parts: (1) weights of each user for every criterion are computed through our proposed modified sub-tree crossover in GP process (2) criteria weights are then incorporated in CF process to generate effective recommendations in our proposed system. The obtained results present significant improvements in prediction and recommendation qualities in comparison to heuristic approaches.
Journal of Statistics and Management Systems, 2019
Facial Gender Identification has vast application in human computer interaction, determines custo... more Facial Gender Identification has vast application in human computer interaction, determines customer profile in shopping centers, and restricted permission to enter in prohibited zone, criminal profile analysis. This paper presents a robust process for illumination invariant compact feature extraction using Gabor filter for the automatic
Load forecasting is a challenging task in power markets that require attention in generating accu... more Load forecasting is a challenging task in power markets that require attention in generating accurate and stable load to deal with planning and management strategies. In past few years, several intelligence-based models have been introduced for precise load forecast. Among them, artificial neural network (ANN) seems more effective and capable to handle the non-linear behavior of load and generates an accurate forecast. However, it suffers from overfitting problem thus reducing the accuracy of load forecasts. To overcome this problem, a hybrid methodology namely ANN-IEAMCGM-R, for short-term load forecast is proposed in this paper. ANN is integrated with an enhanced evolutionary algorithm (IEAMCGM-R) to find optimal network weights. This evolutionary algorithm is composed of improved environmental adaptation method with real parameters (IEAM-R) and our proposed Controlled Gaussian Mutation (CGM) method to bring greater diversity within the population resulting in a higher convergence of solutions. The electric load data from the New England Power Pool (NEPOOL, ISO New England) and Australian Energy Market Operator (New South Wales (NSW), Australia) have been used to illustrate the efficacy of the proposed hybrid methodology. Results show that the proposed hybrid methodology generates higher accuracy than other state-of-theart algorithms.
Collaborative filleting is one of the widely implemented techniques in the area of recommender sy... more Collaborative filleting is one of the widely implemented techniques in the area of recommender systems. But it suffers from data sparsity problem. To address that problem, cross-domain recommender systems (CDRSs) have been emerged to solve the data sparsity problem and improve the accuracy of prediction by transfer learning mechanism. To apply transfer learning mechanism, some common properties associated with users and/or items are needed between the domains. Several attempts have shown that recommendation quality of cross-domain recommender systems could be improved by transferring the user-generated tag information into the target domain. However, sometimes that information is not enough to accomplish recommendation task efficiently. To this end, item features can also be a valuable source of information for developing the correlation between domains and would be considered in generating effective recommendations in target domain. In this paper, we propose a model by utilizing item features and user-generated tags through matrix factorization in CDRSs framework. Firstly, we extract item features in terms of genres and user preferences in terms of user-generated tags. Thereafter, to establish the bridge for transferring knowledge, matrix factorization has been used. Finally, experimental results demonstrate that our proposed model outperforms the other single domain as well as cross domain approaches in CDRSs framework.
In the area of recommender systems, collaborative filtering is widely used technique for recommen... more In the area of recommender systems, collaborative filtering is widely used technique for recommending appropriate items to a user based on the available ratings given by similar users. Most recommender systems (RSs) work only on the single criterion rating i.e., overall rating, however overall rating may not be a good representative of a user preference. Single criterion collaborative filtering (CF) does not generate more reliable recommendations because it suffers from correlation based similarity problems. Moreover, representation of uncertain user preferences is another concern of CF. In our work, we develop a novel fuzzy Bayesian approach to multi-criteria CF for handling uncertain user preferences and correlation based similarity problems. Further, incorporation of multicriteria ratings into CF would be helpful for generating effective recommendations. Through experiments on Yahoo! Movies dataset, we compare our proposed approach to baseline approaches and demonstrate its effectiveness in terms of accuracy, recall and f-measure.
With the rapid advancement of information and communication technologies, e-learning has gained a... more With the rapid advancement of information and communication technologies, e-learning has gained a considerable attention in recent years. Many researchers have attempted to develop various e-learning systems with personalized learning mechanisms for assisting learners so that they can learn more efficiently. In this context, curriculum sequencing is considered as an important concern for developing more efficient personalized e-learning systems. A more effective personalized e-learning recommender system should recommend a sequence of learning materials called learning path, in an appropriate order with a starting and ending point, rather than a sequence of unordered learning materials. Further the recommended sequence should also match the learner preferences for enhancing their learning capabilities. Moreover, the length of recommended sequence cannot be fixed for each learner because these learners differ from one another in their preferences such as knowledge levels, learning styles, emotions, etc. In this paper, we present an effective learning path recommendation system (LPRS) for e-learners through a variable length genetic algorithm (VLGA) by considering learners'
Uploads
Papers by Vibhor Kant