Pub Date : 2019-10-01DOI: 10.1109/ICSSD47982.2019.9002729
Kamal El Guemmat, Sara Ouahabi
Today, search engines play an important role in retrieving documents from a large database. The key success factors of these engines are their indexing and searching techniques.These engines have touched many areas to help their users to find the desired resources in a fast and accurate way. The field of teaching and research take advantage of these engines to offer them to the interested actors (students, teacher, staff, etc.) in order to find the desired learning objects.There are several prominent educational search engines in the field implementing the techniques of indexing and searching either classic, semantic, metadata. However, most engines do not mix all of these to achieve important results.The engine which will be presented in what follows benefits from the best techniques of the literature, and offers a more relevant searching.
{"title":"Towards a new educational search engine based on hybrid searching and indexing techniques","authors":"Kamal El Guemmat, Sara Ouahabi","doi":"10.1109/ICSSD47982.2019.9002729","DOIUrl":"https://doi.org/10.1109/ICSSD47982.2019.9002729","url":null,"abstract":"Today, search engines play an important role in retrieving documents from a large database. The key success factors of these engines are their indexing and searching techniques.These engines have touched many areas to help their users to find the desired resources in a fast and accurate way. The field of teaching and research take advantage of these engines to offer them to the interested actors (students, teacher, staff, etc.) in order to find the desired learning objects.There are several prominent educational search engines in the field implementing the techniques of indexing and searching either classic, semantic, metadata. However, most engines do not mix all of these to achieve important results.The engine which will be presented in what follows benefits from the best techniques of the literature, and offers a more relevant searching.","PeriodicalId":342806,"journal":{"name":"2019 1st International Conference on Smart Systems and Data Science (ICSSD)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132027768","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 : 2019-10-01DOI: 10.1109/ICSSD47982.2019.9002770
Maissae Haddouchi, A. Berrado
Interpretability of highly performant Machine Learning [ML] methods, such as Random Forest [RF], is a key tool that attracts a great interest in datamining research. In the state of the art, RF is well-known as an efficient ensemble learning (in terms of predictive accuracy, flexibility and straightforwardness). Moreover, it is recognized as an intuitive and intelligible approach regarding to its building process. However it is also regarded as a Black Box model because of its hundreds of deep decision trees. This can be crucial for several fields of study, such as healthcare, biology and security, where the lack of interpretability could be a real disadvantage. Indeed, the interpretability of the RF models is, generally, necessary in such fields of applications because of different motivations. In fact, the more the ML users grasp what is going on inside a ML system (process and resulting model), the more they can trust it and take actions based on the knowledge extracted from it. Furthermore, ML models are increasingly constrained by new laws that require regulation and interpretation of the knowledge they provide.Several papers have tackled the interpretation of RF resulting models. It had been associated with different aspects depending on the specificity of the issue studied as well as the users concerned with explanations. Therefore, this paper aims to provide a survey of tools and methods used in literature in order to uncover insights in the RF resulting models. These tools are classified depending on different aspects characterizing the interpretability. This should guide, in practice, in the choice of the most useful tools for interpretation and deep analysis of the RF model depending on the interpretability aspect sought. This should also be valuable for researchers who aim to focus their work on the interpretability of RF, or ML in general.
{"title":"A survey of methods and tools used for interpreting Random Forest","authors":"Maissae Haddouchi, A. Berrado","doi":"10.1109/ICSSD47982.2019.9002770","DOIUrl":"https://doi.org/10.1109/ICSSD47982.2019.9002770","url":null,"abstract":"Interpretability of highly performant Machine Learning [ML] methods, such as Random Forest [RF], is a key tool that attracts a great interest in datamining research. In the state of the art, RF is well-known as an efficient ensemble learning (in terms of predictive accuracy, flexibility and straightforwardness). Moreover, it is recognized as an intuitive and intelligible approach regarding to its building process. However it is also regarded as a Black Box model because of its hundreds of deep decision trees. This can be crucial for several fields of study, such as healthcare, biology and security, where the lack of interpretability could be a real disadvantage. Indeed, the interpretability of the RF models is, generally, necessary in such fields of applications because of different motivations. In fact, the more the ML users grasp what is going on inside a ML system (process and resulting model), the more they can trust it and take actions based on the knowledge extracted from it. Furthermore, ML models are increasingly constrained by new laws that require regulation and interpretation of the knowledge they provide.Several papers have tackled the interpretation of RF resulting models. It had been associated with different aspects depending on the specificity of the issue studied as well as the users concerned with explanations. Therefore, this paper aims to provide a survey of tools and methods used in literature in order to uncover insights in the RF resulting models. These tools are classified depending on different aspects characterizing the interpretability. This should guide, in practice, in the choice of the most useful tools for interpretation and deep analysis of the RF model depending on the interpretability aspect sought. This should also be valuable for researchers who aim to focus their work on the interpretability of RF, or ML in general.","PeriodicalId":342806,"journal":{"name":"2019 1st International Conference on Smart Systems and Data Science (ICSSD)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133247705","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 : 2019-10-01DOI: 10.1109/ICSSD47982.2019.9002857
Imad Sassi, Sara Ouaftouh, S. Anter
Big Data Analytics presents a great opportunity for scientists and businesses. It changed the methods of managing and analyzing the huge amount of data. To make big data valuable, we often use Machine Learning algorithms. Indeed, these algorithms have shown, in the past, their processing speed, efficiency and accuracy. But today, with the complex characteristics of big data, new problems have emerged and we are facing new challenges when developing and designing a new Machine Learning algorithm for Big Data Analytics. Therefore, it is essential to review the classical algorithms to adapt them to this new context. One of the methods of adaptation is the coupling between new technologies (i.e., distributed computing by GPU, Hadoop, Spark) and the Machine Learning algorithms to reduce the computational cost of data analysis. This paper highlights main challenges of adaptation of Machine Learning algorithms to the Big Data context and describes a novel method to make these algorithms efficient and fast in Big Data processing by taking as a case study the Hidden Markov Models using Spark framework. The results of complexity comparison of classical algorithms and those adapted to the Big Data context using Spark show a great improvement.
{"title":"Adaptation of Classical Machine Learning Algorithms to Big Data Context: Problems and Challenges : Case Study: Hidden Markov Models Under Spark","authors":"Imad Sassi, Sara Ouaftouh, S. Anter","doi":"10.1109/ICSSD47982.2019.9002857","DOIUrl":"https://doi.org/10.1109/ICSSD47982.2019.9002857","url":null,"abstract":"Big Data Analytics presents a great opportunity for scientists and businesses. It changed the methods of managing and analyzing the huge amount of data. To make big data valuable, we often use Machine Learning algorithms. Indeed, these algorithms have shown, in the past, their processing speed, efficiency and accuracy. But today, with the complex characteristics of big data, new problems have emerged and we are facing new challenges when developing and designing a new Machine Learning algorithm for Big Data Analytics. Therefore, it is essential to review the classical algorithms to adapt them to this new context. One of the methods of adaptation is the coupling between new technologies (i.e., distributed computing by GPU, Hadoop, Spark) and the Machine Learning algorithms to reduce the computational cost of data analysis. This paper highlights main challenges of adaptation of Machine Learning algorithms to the Big Data context and describes a novel method to make these algorithms efficient and fast in Big Data processing by taking as a case study the Hidden Markov Models using Spark framework. The results of complexity comparison of classical algorithms and those adapted to the Big Data context using Spark show a great improvement.","PeriodicalId":342806,"journal":{"name":"2019 1st International Conference on Smart Systems and Data Science (ICSSD)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114492682","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 : 2019-10-01DOI: 10.1109/ICSSD47982.2019.9002673
Abdelaziz El Yazidi, M. El Kamili, M. L. Hasnaoui
In this article, we present Black SDN, a Software Defined Networking (SDN) architecture for networking and secure communications in wireless sensor networks (WSN). SDN architectures have been developed to improve the routing and networking performance of broadband networks by separating simple controls from data. This basic SDN concept is compatible with WSN networks; However, common SDN implementations for wired networks do not lend themselves directly to distributed mesh networks. SDN promises to improve the lifespan and performance of WSN networks. However, the SDN architecture modifies the WSN network communication schemes, allowing new types of attacks and requiring a new approach to securing the WSN network. Black SDN is a new secure SDN-based network architecture that secures metadata and payload within each layer of a WSN communication packet while using the SDN centralized controller as a trusted third party for secure routing and optimized management of system performance.
{"title":"Black SDN for WSN","authors":"Abdelaziz El Yazidi, M. El Kamili, M. L. Hasnaoui","doi":"10.1109/ICSSD47982.2019.9002673","DOIUrl":"https://doi.org/10.1109/ICSSD47982.2019.9002673","url":null,"abstract":"In this article, we present Black SDN, a Software Defined Networking (SDN) architecture for networking and secure communications in wireless sensor networks (WSN). SDN architectures have been developed to improve the routing and networking performance of broadband networks by separating simple controls from data. This basic SDN concept is compatible with WSN networks; However, common SDN implementations for wired networks do not lend themselves directly to distributed mesh networks. SDN promises to improve the lifespan and performance of WSN networks. However, the SDN architecture modifies the WSN network communication schemes, allowing new types of attacks and requiring a new approach to securing the WSN network. Black SDN is a new secure SDN-based network architecture that secures metadata and payload within each layer of a WSN communication packet while using the SDN centralized controller as a trusted third party for secure routing and optimized management of system performance.","PeriodicalId":342806,"journal":{"name":"2019 1st International Conference on Smart Systems and Data Science (ICSSD)","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114874045","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 : 2019-10-01DOI: 10.1109/ICSSD47982.2019.9002820
Morad Hajji, Mohammed Qbadou, K. Mansouri
The Extract-Transform-Load (ETL) process is the most widely used mechanism to keep a Data Warehouse loading with data extracted from a variety of sources. Currently, tools offering graphical interfaces to facilitate the manipulation of ETL processes have become very popular and have reached a very advanced level of maturity. Talend Open Studio for Data Integration is one of the most popular and comprehensive tools in terms of functionality and performance. So far, this ETL tool provides a large number of components for different data sources. However, the advent of the Semantic Web brings the notion of ontology as a new source of data whose structure is characterized by its complex aspect related to the expressiveness of languages of the knowledge representation. The emergence of this notion is a new challenge. Indeed, to our knowledge, Talend Open Studio for Data Integration does not have any components intended to support ontological sources.In this contribution, we present our approach for the development of Talend Open Studio for Data Integration components in order to use Semantic Web data in ETL processes created with this tool. Using a strategy that promotes the abstraction of ontological sources, this approach can be adapted to different languages of representation of knowledge such as RDF and OWL.In order to assess the usefulness of our approach, we evaluated it as part of a hypothetical example set of a simplistic ontology.
提取-转换-加载(Extract-Transform-Load, ETL)过程是使用最广泛的机制，用于保持数据仓库加载从各种来源提取的数据。目前，提供图形界面以方便操作ETL过程的工具已经变得非常流行，并且已经达到了非常高级的成熟度。Talend Open Studio for Data Integration是在功能和性能方面最流行和最全面的工具之一。到目前为止，这个ETL工具为不同的数据源提供了大量的组件。然而，语义网的出现带来了本体作为一种新的数据源的概念，其结构的特点是其复杂性与知识表示语言的表达性有关。这个概念的出现是一个新的挑战。事实上，据我们所知，Talend Open Studio for Data Integration并没有任何支持本体论源的组件。在本文中，我们介绍了开发Talend Open Studio for Data Integration组件的方法，以便在使用该工具创建的ETL流程中使用语义Web数据。使用一种促进本体论源抽象的策略，这种方法可以适应不同的知识表示语言，如RDF和OWL。为了评估我们的方法的有用性，我们将其作为一个简单本体的假设示例集的一部分进行评估。
{"title":"Towards the Development of Talend Open Studio Components for the Support of Semantic Sources","authors":"Morad Hajji, Mohammed Qbadou, K. Mansouri","doi":"10.1109/ICSSD47982.2019.9002820","DOIUrl":"https://doi.org/10.1109/ICSSD47982.2019.9002820","url":null,"abstract":"The Extract-Transform-Load (ETL) process is the most widely used mechanism to keep a Data Warehouse loading with data extracted from a variety of sources. Currently, tools offering graphical interfaces to facilitate the manipulation of ETL processes have become very popular and have reached a very advanced level of maturity. Talend Open Studio for Data Integration is one of the most popular and comprehensive tools in terms of functionality and performance. So far, this ETL tool provides a large number of components for different data sources. However, the advent of the Semantic Web brings the notion of ontology as a new source of data whose structure is characterized by its complex aspect related to the expressiveness of languages of the knowledge representation. The emergence of this notion is a new challenge. Indeed, to our knowledge, Talend Open Studio for Data Integration does not have any components intended to support ontological sources.In this contribution, we present our approach for the development of Talend Open Studio for Data Integration components in order to use Semantic Web data in ETL processes created with this tool. Using a strategy that promotes the abstraction of ontological sources, this approach can be adapted to different languages of representation of knowledge such as RDF and OWL.In order to assess the usefulness of our approach, we evaluated it as part of a hypothetical example set of a simplistic ontology.","PeriodicalId":342806,"journal":{"name":"2019 1st International Conference on Smart Systems and Data Science (ICSSD)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125296404","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 : 2019-10-01DOI: 10.1109/ICSSD47982.2019.9002867
Maryame Naji, Daoudi Najima, Rahimi Hasnae, R. Ajhoun
With the explosion of web 2.0, we are witnessing a sharp increase in Internet users such as a vertiginous evolution of social media. These Media constitute a source of rich and varied information for researchers in sentiment analysis. In this paper, we introduce a method to build a Dataset by extracting and processing Data from Twitter. In fact, we present in this paper, a novel approach to represent our Dataset the way to improve the prediction of profiles with depression by considering social media factors as analysis parameters.
{"title":"Customized data extraction and processing for the prediction of Baby Blues from social media","authors":"Maryame Naji, Daoudi Najima, Rahimi Hasnae, R. Ajhoun","doi":"10.1109/ICSSD47982.2019.9002867","DOIUrl":"https://doi.org/10.1109/ICSSD47982.2019.9002867","url":null,"abstract":"With the explosion of web 2.0, we are witnessing a sharp increase in Internet users such as a vertiginous evolution of social media. These Media constitute a source of rich and varied information for researchers in sentiment analysis. In this paper, we introduce a method to build a Dataset by extracting and processing Data from Twitter. In fact, we present in this paper, a novel approach to represent our Dataset the way to improve the prediction of profiles with depression by considering social media factors as analysis parameters.","PeriodicalId":342806,"journal":{"name":"2019 1st International Conference on Smart Systems and Data Science (ICSSD)","volume":"95 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126330888","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 : 2019-10-01DOI: 10.1109/ICSSD47982.2019.9003112
Majdouline Meddad, Chouaib Moujahdi, M. Mikram, M. Rziza
Computer vision is a field that handles how a computer can gain a high level of understanding from several kind of inputted information. In general, it tries to automate tasks that humans have the ability to do. Computer vision tasks contain a method to understand digital images and extract high dimensional in order to generate a piece of symbolic information, for example, in the forms of decisions. One of the open issues in image processing and computer vision nowadays is how we can handle the problem of quick identification in a multi-user identification system. In this paper, We propose an identification system in a big data environment that provide and an acceptance identification time while keeping a good performance, in uncontrolled conditions, in comparison with some compared classical systems.
{"title":"A Multi-User Face Identification system with distributed tasks in a Big Data Environment","authors":"Majdouline Meddad, Chouaib Moujahdi, M. Mikram, M. Rziza","doi":"10.1109/ICSSD47982.2019.9003112","DOIUrl":"https://doi.org/10.1109/ICSSD47982.2019.9003112","url":null,"abstract":"Computer vision is a field that handles how a computer can gain a high level of understanding from several kind of inputted information. In general, it tries to automate tasks that humans have the ability to do. Computer vision tasks contain a method to understand digital images and extract high dimensional in order to generate a piece of symbolic information, for example, in the forms of decisions. One of the open issues in image processing and computer vision nowadays is how we can handle the problem of quick identification in a multi-user identification system. In this paper, We propose an identification system in a big data environment that provide and an acceptance identification time while keeping a good performance, in uncontrolled conditions, in comparison with some compared classical systems.","PeriodicalId":342806,"journal":{"name":"2019 1st International Conference on Smart Systems and Data Science (ICSSD)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131732184","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 : 2019-10-01DOI: 10.1109/ICSSD47982.2019.9003097
Rym Nassih, A. Berrado
This paper is an overview of the bump hunting algorithm and in particular the Patient Rule Induction Method and its applications. We also give an overview about interpretability in several key supervised data mining algorithms. This allows for exploring the potential for using PRIM, with its interpretation capability, as a core technology towards building a highly accurate and interpretable classifier in a mixed data space.
{"title":"Towards a patient rule induction method based classifier","authors":"Rym Nassih, A. Berrado","doi":"10.1109/ICSSD47982.2019.9003097","DOIUrl":"https://doi.org/10.1109/ICSSD47982.2019.9003097","url":null,"abstract":"This paper is an overview of the bump hunting algorithm and in particular the Patient Rule Induction Method and its applications. We also give an overview about interpretability in several key supervised data mining algorithms. This allows for exploring the potential for using PRIM, with its interpretation capability, as a core technology towards building a highly accurate and interpretable classifier in a mixed data space.","PeriodicalId":342806,"journal":{"name":"2019 1st International Conference on Smart Systems and Data Science (ICSSD)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120971166","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 : 2019-10-01DOI: 10.1109/ICSSD47982.2019.9003165
Nihal Elkhalidi, F. Benabbou, N. Sael
Finding a parking spot in a big city is often frustrating for drivers. The search for a parking space can waste a lot of time, that’s why, it is necessary to adopt a good management of car parks in all cities. In recent years, several studies have proposed intelligent parking management systems using parking reservation techniques to guarantee an optimal parking space for drivers meeting their needs. We have proposed in a previous work a distributed parking management system based on multiagent systems where we have explained the different agents ensuring its operation. The reservation agent is an integral part of this system. In this article, we propose a state of the art on the different techniques used in the reservation systems. the aim is to analyze the different solutions proposed in this context and make a comparison according to different criteria such as competitive reservation, the respect of constraints, scalability, optimization, performance… The objective of this study is to highlight the strengths and weaknesses of each technique in order to propose a complete optimization model on which will be based our reservation agent.
{"title":"Survey of Reservation Techniques in Smart Parking","authors":"Nihal Elkhalidi, F. Benabbou, N. Sael","doi":"10.1109/ICSSD47982.2019.9003165","DOIUrl":"https://doi.org/10.1109/ICSSD47982.2019.9003165","url":null,"abstract":"Finding a parking spot in a big city is often frustrating for drivers. The search for a parking space can waste a lot of time, that’s why, it is necessary to adopt a good management of car parks in all cities. In recent years, several studies have proposed intelligent parking management systems using parking reservation techniques to guarantee an optimal parking space for drivers meeting their needs. We have proposed in a previous work a distributed parking management system based on multiagent systems where we have explained the different agents ensuring its operation. The reservation agent is an integral part of this system. In this article, we propose a state of the art on the different techniques used in the reservation systems. the aim is to analyze the different solutions proposed in this context and make a comparison according to different criteria such as competitive reservation, the respect of constraints, scalability, optimization, performance… The objective of this study is to highlight the strengths and weaknesses of each technique in order to propose a complete optimization model on which will be based our reservation agent.","PeriodicalId":342806,"journal":{"name":"2019 1st International Conference on Smart Systems and Data Science (ICSSD)","volume":"66 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130951651","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 : 2019-10-01DOI: 10.1109/ICSSD47982.2019.9003154
Khadija Elghomary, D. Bouzidi
The choice of the relevant partners to interact and collaborate is a sensitive issue, especially in open, dynamic and heterogeneous environments such as the MOOC platforms marked by a large number of learners having different needs and expectations. In these contexts, the process of detecting the trusted user to interact and collaborate is more difficult and time-consuming which can be contributor to learner’s demotivation and disengagement.Trust models could be efficiently adopted in these platforms to increase completion rates and to boost learner’s motivation by helping them to find the appropriate partner (peer) for meeting their needs and achieving their learning objectives. In general, several methods used to recommend peers in MOOCs are based particularly on similarity between learners profiles without considering the dynamicity of their behaviors and interests, or the influence of the trust relationships among them which impact strongly the selection of the suitable partner.In this paper we provide architecture of Dynamic Peer Recommendation System (DPRS) based on trust management system (TMS) that represent an adaptation and inspiration of one of the major and recent works of dynamic SIoT trust models in order to adjust to the high level of dynamism and mobility of MOOCs. This architecture eases the decision making process to select relevant partner by MOOC learners to guarantee an engaging learning experience and promote a peer-collaboration in this community.
{"title":"Dynamic Peer Recommendation System based on Trust Model for sustainable social tutoring in MOOCs","authors":"Khadija Elghomary, D. Bouzidi","doi":"10.1109/ICSSD47982.2019.9003154","DOIUrl":"https://doi.org/10.1109/ICSSD47982.2019.9003154","url":null,"abstract":"The choice of the relevant partners to interact and collaborate is a sensitive issue, especially in open, dynamic and heterogeneous environments such as the MOOC platforms marked by a large number of learners having different needs and expectations. In these contexts, the process of detecting the trusted user to interact and collaborate is more difficult and time-consuming which can be contributor to learner’s demotivation and disengagement.Trust models could be efficiently adopted in these platforms to increase completion rates and to boost learner’s motivation by helping them to find the appropriate partner (peer) for meeting their needs and achieving their learning objectives. In general, several methods used to recommend peers in MOOCs are based particularly on similarity between learners profiles without considering the dynamicity of their behaviors and interests, or the influence of the trust relationships among them which impact strongly the selection of the suitable partner.In this paper we provide architecture of Dynamic Peer Recommendation System (DPRS) based on trust management system (TMS) that represent an adaptation and inspiration of one of the major and recent works of dynamic SIoT trust models in order to adjust to the high level of dynamism and mobility of MOOCs. This architecture eases the decision making process to select relevant partner by MOOC learners to guarantee an engaging learning experience and promote a peer-collaboration in this community.","PeriodicalId":342806,"journal":{"name":"2019 1st International Conference on Smart Systems and Data Science (ICSSD)","volume":"80 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134278561","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：