Pub Date : 2019-07-15DOI: 10.1109/IISA.2019.8900678
Dionisis Margaris, Dionysios Vasilopoulos, C. Vassilakis, D. Spiliotopoulos
Collaborative filtering creates personalized recommendations by considering ratings entered by users. Collaborative filtering algorithms initially detect users whose likings are alike, by exploring the similarity between ratings that have insofar been submitted. Users having a high degree of similarity regarding their ratings are termed near neighbors, and in order to formulate a recommendation for a user, her near neighbors’ ratings are extracted and form the basis for the recommendation. Collaborative filtering algorithms however exhibit the problem commonly referred to as “gray sheep this pertains to the case where for some users no near neighbors can be identified, and hence no personalized recommendations can be computed. The “gray sheep” problem is more severe in sparse datasets, i.e. datasets where the number of ratings is small, compared to the number of items and users. In this paper, we address the “gray sheep” problem by introducing the concept of virtual near neighbors and a related algorithm for their creation on the basis of the existing ones. We evaluate the proposed algorithm, which is termed as CFVNN, using eight widely used datasets and considering two correlation metrics which are widely used in Collaborative Filtering research, namely the Pearson Correlation Coefficient and the Cosine Similarity. The results show that the proposed algorithm considerably leverages the capability of a Collaborative Filtering system to compute personalized recommendations in the context of sparse datasets, tackling thus efficiently the “gray sheep” problem. In parallel, the CFVNN algorithm achieves improvements in rating prediction quality, as this is expressed through the Mean Absolute Error and the Root Mean Square Error metrics.
协同过滤通过考虑用户输入的评分创建个性化推荐。协同过滤算法最初通过探索迄今为止提交的评分之间的相似性来检测喜欢相似的用户。在评分方面具有高度相似性的用户被称为近邻用户，为了制定对用户的推荐，提取其近邻的评分并形成推荐的基础。然而，协同过滤算法显示了通常被称为“灰羊”的问题，这涉及到某些用户无法识别近邻，因此无法计算个性化推荐的情况。“灰羊”问题在稀疏数据集中更为严重，即与项目和用户数量相比，评级数量较少的数据集。在本文中，我们通过引入虚拟近邻的概念以及在现有近邻的基础上创建虚拟近邻的相关算法来解决“灰羊”问题。我们使用8个广泛使用的数据集，并考虑在协同过滤研究中广泛使用的两个相关度量，即Pearson相关系数和余弦相似度，来评估所提出的CFVNN算法。结果表明，该算法充分利用了协同过滤系统在稀疏数据集背景下计算个性化推荐的能力，从而有效地解决了“灰羊”问题。与此同时，CFVNN算法在评级预测质量上取得了进步，因为这是通过Mean Absolute Error和Root Mean Square Error度量来表达的。
{"title":"Improving Collaborative Filtering’s Rating Prediction Coverage in Sparse Datasets through the Introduction of Virtual Near Neighbors","authors":"Dionisis Margaris, Dionysios Vasilopoulos, C. Vassilakis, D. Spiliotopoulos","doi":"10.1109/IISA.2019.8900678","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900678","url":null,"abstract":"Collaborative filtering creates personalized recommendations by considering ratings entered by users. Collaborative filtering algorithms initially detect users whose likings are alike, by exploring the similarity between ratings that have insofar been submitted. Users having a high degree of similarity regarding their ratings are termed near neighbors, and in order to formulate a recommendation for a user, her near neighbors’ ratings are extracted and form the basis for the recommendation. Collaborative filtering algorithms however exhibit the problem commonly referred to as “gray sheep this pertains to the case where for some users no near neighbors can be identified, and hence no personalized recommendations can be computed. The “gray sheep” problem is more severe in sparse datasets, i.e. datasets where the number of ratings is small, compared to the number of items and users. In this paper, we address the “gray sheep” problem by introducing the concept of virtual near neighbors and a related algorithm for their creation on the basis of the existing ones. We evaluate the proposed algorithm, which is termed as CFVNN, using eight widely used datasets and considering two correlation metrics which are widely used in Collaborative Filtering research, namely the Pearson Correlation Coefficient and the Cosine Similarity. The results show that the proposed algorithm considerably leverages the capability of a Collaborative Filtering system to compute personalized recommendations in the context of sparse datasets, tackling thus efficiently the “gray sheep” problem. In parallel, the CFVNN algorithm achieves improvements in rating prediction quality, as this is expressed through the Mean Absolute Error and the Root Mean Square Error metrics.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129960067","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-15DOI: 10.1109/IISA.2019.8900775
K. Chrysafiadi, M. Virvou
In this paper a novel fuzzy mechanism for automatic reasoning of human errors in operating systems is presented. The presented mechanism combines Human Plausible Reasoning (HPR) theory with fuzzy logic. HPR is used for inferring the commands the user of an operating system should have type and fuzzy logic is used to handle the uncertainty that characterizes the complex reasoning process by modelling the errors’ types in a more realistic way. Particularly, the output of HPR theory guesses all the possible command that the user may wants to type. These guesses can include many types of errors varying from typographic to wrong use of a legal command. In the presented mechanism, these guesses are input in a fuzzy reasoner, which takes into account the needs, characteristics and misconceptions of each individual user and decides about the most appropriate explanation of user’s error and gives personalized advice that fits better in each context and situation. The mechanism has been applied on the sub-domain of file manipulation of UNIX. The potential of the presented mechanism to reason about operating system’s users’ slips and misconceptions are discussed.
{"title":"Enhancing Automatic Reasoning of human errors in an operating system using fuzzy logic","authors":"K. Chrysafiadi, M. Virvou","doi":"10.1109/IISA.2019.8900775","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900775","url":null,"abstract":"In this paper a novel fuzzy mechanism for automatic reasoning of human errors in operating systems is presented. The presented mechanism combines Human Plausible Reasoning (HPR) theory with fuzzy logic. HPR is used for inferring the commands the user of an operating system should have type and fuzzy logic is used to handle the uncertainty that characterizes the complex reasoning process by modelling the errors’ types in a more realistic way. Particularly, the output of HPR theory guesses all the possible command that the user may wants to type. These guesses can include many types of errors varying from typographic to wrong use of a legal command. In the presented mechanism, these guesses are input in a fuzzy reasoner, which takes into account the needs, characteristics and misconceptions of each individual user and decides about the most appropriate explanation of user’s error and gives personalized advice that fits better in each context and situation. The mechanism has been applied on the sub-domain of file manipulation of UNIX. The potential of the presented mechanism to reason about operating system’s users’ slips and misconceptions are discussed.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"22 8","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132090948","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-15DOI: 10.1109/IISA.2019.8900729
C. Troussas, Akrivi Krouska, M. Virvou
Mobile learning uncovers new dimensions of learning and personal growth. Mobile phones have completely dominated our lives from communication and entertainment to socializing and learning. In view of providing more individualized learning through mobile phones, several intelligent techniques should be incorporated in mobile-assisted learning systems. As such, this paper presents an effective analysis of students’ errors during the assessment process in mobile learning using Natural Language Processing (NLP) techniques. The error analysis can reason between grammatical, syntax and careless errors using the Levenshtein distance. Moreover, it describes dynamic methods for motivating students in order to improve their learning experience. As such, students can receive motivation in case of making errors, cognitive inconsistencies, etc. Dynamic motivation is enriched with the delivery of badges as a means to further enhance knowledge acquisition. As a testbed for our research, a mobile language learning application for tutoring the English language has been designed, fully developed and evaluated. Concluding, this paper presents real examples of operation of the presented system and the evaluation results show the acceptance of the NLP-based error analysis and the dynamic motivation techniques by students.
{"title":"NLP-based error analysis and dynamic motivation techniques in mobile learning","authors":"C. Troussas, Akrivi Krouska, M. Virvou","doi":"10.1109/IISA.2019.8900729","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900729","url":null,"abstract":"Mobile learning uncovers new dimensions of learning and personal growth. Mobile phones have completely dominated our lives from communication and entertainment to socializing and learning. In view of providing more individualized learning through mobile phones, several intelligent techniques should be incorporated in mobile-assisted learning systems. As such, this paper presents an effective analysis of students’ errors during the assessment process in mobile learning using Natural Language Processing (NLP) techniques. The error analysis can reason between grammatical, syntax and careless errors using the Levenshtein distance. Moreover, it describes dynamic methods for motivating students in order to improve their learning experience. As such, students can receive motivation in case of making errors, cognitive inconsistencies, etc. Dynamic motivation is enriched with the delivery of badges as a means to further enhance knowledge acquisition. As a testbed for our research, a mobile language learning application for tutoring the English language has been designed, fully developed and evaluated. Concluding, this paper presents real examples of operation of the presented system and the evaluation results show the acceptance of the NLP-based error analysis and the dynamic motivation techniques by students.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122177517","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-15DOI: 10.1109/IISA.2019.8900761
Argyro Tsirtsi, E. Zacharaki, Spyridon Kalogiannis, V. Megalooikonomou
The last years there is a great interest in developing unobtrusive health monitoring systems with a predictive component, aiming to recognize signs of illness in an attempt to assist clinicians in delivering early interventions. The objective of this work is to investigate whether the physiological and kinetic functioning and human activity of daily living monitored by multiple sensors can be used as surrogate of the standard clinical assessment. We focus on the older population and propose to utilize Multiple Instance Learning (MIL) to predict their clinical profile from the multi-sensorial data. ReliefF-MI is applied to achieve dimensionality reduction and to discover the most important features that are associated with each clinical metric, while the BagSMOTE algorithm is utilized to mitigate the class imbalance problem. The proposed methodology was evaluated on a multi-parametric dataset of 86 older adults containing clinical parameters from various domains (cognitive, physical, medical, psychological, social and showed high prognostic capacity for the person’s functionality (Katz index) and social interaction (phone calls).
{"title":"Clinical profile prediction by multiple instance learning from multi-sensorial data","authors":"Argyro Tsirtsi, E. Zacharaki, Spyridon Kalogiannis, V. Megalooikonomou","doi":"10.1109/IISA.2019.8900761","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900761","url":null,"abstract":"The last years there is a great interest in developing unobtrusive health monitoring systems with a predictive component, aiming to recognize signs of illness in an attempt to assist clinicians in delivering early interventions. The objective of this work is to investigate whether the physiological and kinetic functioning and human activity of daily living monitored by multiple sensors can be used as surrogate of the standard clinical assessment. We focus on the older population and propose to utilize Multiple Instance Learning (MIL) to predict their clinical profile from the multi-sensorial data. ReliefF-MI is applied to achieve dimensionality reduction and to discover the most important features that are associated with each clinical metric, while the BagSMOTE algorithm is utilized to mitigate the class imbalance problem. The proposed methodology was evaluated on a multi-parametric dataset of 86 older adults containing clinical parameters from various domains (cognitive, physical, medical, psychological, social and showed high prognostic capacity for the person’s functionality (Katz index) and social interaction (phone calls).","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126103709","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-15DOI: 10.1109/IISA.2019.8900675
N. Gorgolis, I. Hatzilygeroudis, Z. Istenes, Lazlo-Grad Gyenne
Next word prediction is an important problem in the domain of NLP, hence in modern artificial intelligence. It draws both scientific and industrial interest, as it consists the core of many processes, like autocorrection, text generation, review prediction etc. Currently, the most efficient and common approach used is classification, using artificial neural networks (ANNs). One of the main drawbacks of ANNs is fine – tuning their hyperparameters, a procedure which is essential to the performance of the model. On the other hand, the approaches usually used for fine – tuning are either computationally unaffordable (e.g. grid search) or of uncertain efficiency (e.g. trial & error). As a response to the above, through the current paper is presented a simple genetic algorithm approach, which is used for the hyperparameter tuning of a common language model and it achieves tuning efficiency without following an exhaustive search.
{"title":"Hyperparameter Optimization of LSTM Network Models through Genetic Algorithm","authors":"N. Gorgolis, I. Hatzilygeroudis, Z. Istenes, Lazlo-Grad Gyenne","doi":"10.1109/IISA.2019.8900675","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900675","url":null,"abstract":"Next word prediction is an important problem in the domain of NLP, hence in modern artificial intelligence. It draws both scientific and industrial interest, as it consists the core of many processes, like autocorrection, text generation, review prediction etc. Currently, the most efficient and common approach used is classification, using artificial neural networks (ANNs). One of the main drawbacks of ANNs is fine – tuning their hyperparameters, a procedure which is essential to the performance of the model. On the other hand, the approaches usually used for fine – tuning are either computationally unaffordable (e.g. grid search) or of uncertain efficiency (e.g. trial & error). As a response to the above, through the current paper is presented a simple genetic algorithm approach, which is used for the hyperparameter tuning of a common language model and it achieves tuning efficiency without following an exhaustive search.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125084531","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-15DOI: 10.1109/IISA.2019.8900724
Stamatis Karlos, V. G. Kanas, Christos K. Aridas, Nikos Fazakis, S. Kotsiantis
In real-world cases, handling of both labeled and unlabeled data has raised the interest of several data scientists and Machine Learning engineers, leading to several demonstrations that apply data augmenting approaches to achieve an effective learning behavior. Although the majority of them propose either the exploitation of Semi-supervised or Active Learning approaches, individually, their combination has not been widely used. The ambition of this strategy is the efficient utilization of the available human knowledge relying along with the decisions driven by automated methods under a common framework. Thus, we conduct an empirical evaluation of such a combinatory approach over three problems, related to multimodal data operating under the pool-based scenario: Gender Identification, Recognition of Offensive Language and Emotion Detection. Into the proposed learning framework, which exploits initially labeled instances with small cardinality, our results prove the benefits of adopting such kind of semi-automated approaches regarding both the achieved predictive correctness and the reduced consumption of time and cost resources, as well as the smoothness of the learning convergence, mainly using ensemble classifiers.
{"title":"Combining Active Learning with Self-train algorithm for classification of multimodal problems","authors":"Stamatis Karlos, V. G. Kanas, Christos K. Aridas, Nikos Fazakis, S. Kotsiantis","doi":"10.1109/IISA.2019.8900724","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900724","url":null,"abstract":"In real-world cases, handling of both labeled and unlabeled data has raised the interest of several data scientists and Machine Learning engineers, leading to several demonstrations that apply data augmenting approaches to achieve an effective learning behavior. Although the majority of them propose either the exploitation of Semi-supervised or Active Learning approaches, individually, their combination has not been widely used. The ambition of this strategy is the efficient utilization of the available human knowledge relying along with the decisions driven by automated methods under a common framework. Thus, we conduct an empirical evaluation of such a combinatory approach over three problems, related to multimodal data operating under the pool-based scenario: Gender Identification, Recognition of Offensive Language and Emotion Detection. Into the proposed learning framework, which exploits initially labeled instances with small cardinality, our results prove the benefits of adopting such kind of semi-automated approaches regarding both the achieved predictive correctness and the reduced consumption of time and cost resources, as well as the smoothness of the learning convergence, mainly using ensemble classifiers.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131299518","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-15DOI: 10.1109/IISA.2019.8900699
Nikolaos Spatiotis, I. Perikos, I. Mporas, M. Paraskevas
Nowadays, information, communication and interaction between people worldwide have been facilitated by the rapid development of technology and they are mainly achieved through the internet. Internet users are now new creators of information data and express their ideas, their opinions, their feelings and their attitudes about products and services rather than passive information recipients. Given the evolution of modern technological advances, such as the proliferation of mobile devices social networks and services is extending. User-generated content in social media constitutes a very meaningful information source and consists of opinions towards various events and services. In this paper, we present a methodology that aims to analyze Greek text and extract indicative info towards users’ opinions and attitudes. Specifically, we describe a supervised approach adopted that analyzes and classifies comments and reviews into the appropriate polarity category. Discretization techniques are also applied to improve the performance and the accuracy of classification procedures. Finally, we present an experimental evaluation that was designed and conducted and which revealed quite interesting findings.
{"title":"Examining the Impact of Discretization Technique on Sentiment Analysis for the Greek Language","authors":"Nikolaos Spatiotis, I. Perikos, I. Mporas, M. Paraskevas","doi":"10.1109/IISA.2019.8900699","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900699","url":null,"abstract":"Nowadays, information, communication and interaction between people worldwide have been facilitated by the rapid development of technology and they are mainly achieved through the internet. Internet users are now new creators of information data and express their ideas, their opinions, their feelings and their attitudes about products and services rather than passive information recipients. Given the evolution of modern technological advances, such as the proliferation of mobile devices social networks and services is extending. User-generated content in social media constitutes a very meaningful information source and consists of opinions towards various events and services. In this paper, we present a methodology that aims to analyze Greek text and extract indicative info towards users’ opinions and attitudes. Specifically, we describe a supervised approach adopted that analyzes and classifies comments and reviews into the appropriate polarity category. Discretization techniques are also applied to improve the performance and the accuracy of classification procedures. Finally, we present an experimental evaluation that was designed and conducted and which revealed quite interesting findings.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128457406","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-15DOI: 10.1109/IISA.2019.8900743
Andrea Mussati, Christian Giang, Alberto Piatti, F. Mondada
In the past, the use of tangible programming languages has shown several advantages compared to screenbased graphical programming languages. Especially when presented to novices, such interfaces may represent a more intuitive and straightforward alternative to teach basic computer science and programming concepts. Previous studies have reported increased interest and improved collaboration when tangible programming languages were used. However, additional financial expenses have often hindered the use of such interfaces in formal education settings. This work therefore presents a low-cost and customizable solution of a tangible programming language for Thymio, an educational robot widely used in primary and secondary schools. Using a computer vision algorithm, graphical icons printed on paper are captured by a camera, and subsequently interpreted and sent to the robot for execution. Two user studies with in total 77 university students showed promising results, indicating that the devised interface can elicit more interest and a higher level of collaboration within groups.
{"title":"A Tangible Programming Language for the Educational Robot Thymio","authors":"Andrea Mussati, Christian Giang, Alberto Piatti, F. Mondada","doi":"10.1109/IISA.2019.8900743","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900743","url":null,"abstract":"In the past, the use of tangible programming languages has shown several advantages compared to screenbased graphical programming languages. Especially when presented to novices, such interfaces may represent a more intuitive and straightforward alternative to teach basic computer science and programming concepts. Previous studies have reported increased interest and improved collaboration when tangible programming languages were used. However, additional financial expenses have often hindered the use of such interfaces in formal education settings. This work therefore presents a low-cost and customizable solution of a tangible programming language for Thymio, an educational robot widely used in primary and secondary schools. Using a computer vision algorithm, graphical icons printed on paper are captured by a camera, and subsequently interpreted and sent to the robot for execution. Two user studies with in total 77 university students showed promising results, indicating that the devised interface can elicit more interest and a higher level of collaboration within groups.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126491551","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-15DOI: 10.1109/IISA.2019.8900758
Dionisis Margaris, Dionysios Vasilopoulos, C. Vassilakis, D. Spiliotopoulos
Collaborative filtering formulates personalized recommendations by considering ratings submitted by users. Collaborative filtering algorithms initially find people having similar likings, by inspecting the similarity of ratings already present in the ratings database. Users exhibiting high similarity regarding their likings are classified as “near neighbors” (NNs) and the ratings entered by each user’s near neighbors drive the formulation of recommendations for that user. To quantify the similarity between users, in order to determine a user’s NNs, a similarity metric is used. Insofar, similarity metrics proposed in the literature either consider all user ratings equally or take into account temporal variations within the users’ or items’ ratings history. However users’ ratings are co-shaped according to the experiences that they had in the past; therefore if two users enter similar (or dissimilar) ratings for an item while having experienced to a large extent the same items in the past, this constitutes stronger evidence about user similarity (or dissimilarity). Insofar however, no similarity metric takes into account this aspect. In this work, we propose and evaluate an algorithm that considers the common item rating past when computing rating predictions, in order to increase rating prediction accuracy.
{"title":"Improving Collaborative Filtering’s Rating Prediction Accuracy by Introducing the Common Item Rating Past Criterion","authors":"Dionisis Margaris, Dionysios Vasilopoulos, C. Vassilakis, D. Spiliotopoulos","doi":"10.1109/IISA.2019.8900758","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900758","url":null,"abstract":"Collaborative filtering formulates personalized recommendations by considering ratings submitted by users. Collaborative filtering algorithms initially find people having similar likings, by inspecting the similarity of ratings already present in the ratings database. Users exhibiting high similarity regarding their likings are classified as “near neighbors” (NNs) and the ratings entered by each user’s near neighbors drive the formulation of recommendations for that user. To quantify the similarity between users, in order to determine a user’s NNs, a similarity metric is used. Insofar, similarity metrics proposed in the literature either consider all user ratings equally or take into account temporal variations within the users’ or items’ ratings history. However users’ ratings are co-shaped according to the experiences that they had in the past; therefore if two users enter similar (or dissimilar) ratings for an item while having experienced to a large extent the same items in the past, this constitutes stronger evidence about user similarity (or dissimilarity). Insofar however, no similarity metric takes into account this aspect. In this work, we propose and evaluate an algorithm that considers the common item rating past when computing rating predictions, in order to increase rating prediction accuracy.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133890117","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-15DOI: 10.1109/IISA.2019.8900737
Nikos Fazakis, Georgios Kostopoulos, Stamatis Karlos, S. Kotsiantis, K. Sgarbas
Semi-Supervised Learning (SSL) is an ever-growing research area offering a powerful set of methods, either single or multi-view, for exploiting both labeled and unlabeled instances in the most effective manner. Self-training is a representative SSL algorithm which has been efficiently implemented for solving several classification problems in a wide range of scientific fields. Moreover, self-training has served as the base for the development of several self-labeled methods. In addition, gradient boosting is an advanced machine learning technique, a boosting algorithm for both classification and regression problems, which produces a predictive model in the form of decision trees. In this context, the principal objective of this paper is to put forward an improved self-training algorithm for classification tasks utilizing the efficacy of eXtreme Gradient Boosting (XGBoost) trees in a self-labeled scheme in order to build a highly accurate and robust classification model. A number of experiments on benchmark datasets were executed demonstrating the superiority of the proposed method over representative semi-supervised methods, as statistically verified by the Friedman non-parametric test.
{"title":"Self-trained eXtreme Gradient Boosting Trees","authors":"Nikos Fazakis, Georgios Kostopoulos, Stamatis Karlos, S. Kotsiantis, K. Sgarbas","doi":"10.1109/IISA.2019.8900737","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900737","url":null,"abstract":"Semi-Supervised Learning (SSL) is an ever-growing research area offering a powerful set of methods, either single or multi-view, for exploiting both labeled and unlabeled instances in the most effective manner. Self-training is a representative SSL algorithm which has been efficiently implemented for solving several classification problems in a wide range of scientific fields. Moreover, self-training has served as the base for the development of several self-labeled methods. In addition, gradient boosting is an advanced machine learning technique, a boosting algorithm for both classification and regression problems, which produces a predictive model in the form of decision trees. In this context, the principal objective of this paper is to put forward an improved self-training algorithm for classification tasks utilizing the efficacy of eXtreme Gradient Boosting (XGBoost) trees in a self-labeled scheme in order to build a highly accurate and robust classification model. A number of experiments on benchmark datasets were executed demonstrating the superiority of the proposed method over representative semi-supervised methods, as statistically verified by the Friedman non-parametric test.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125643292","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：