Pub Date : 2016-10-24DOI: 10.1109/CIST.2016.7805076
Abdelkader El Mahdaouy, Said Ouatik El Alaoui, Éric Gaussier
Traditional Information Retrieval (IR) models are based on bag-of-words paradigm, where relevance scores are computed based on exact matching of keywords. Although these models have already achieved good performance, it has been shown that most of dissatisfaction cases in relevance are due to term mismatch between queries and documents. In this paper, we introduce novel method to compute term frequency based on semantic similarities using distributed representations of words in a vector space (Word Embeddings). Our main goal is to allow distinct but semantically related terms to match each other and contribute to the relevance scores. Hence, Arabic documents are retrieved beyond the bag-of-words paradigm based on semantic similarities between word vectors. The results on Arabic standard TREC data sets show significant improvement over the baseline bag-of-words models.
{"title":"Semantically enhanced term frequency based on word embeddings for Arabic information retrieval","authors":"Abdelkader El Mahdaouy, Said Ouatik El Alaoui, Éric Gaussier","doi":"10.1109/CIST.2016.7805076","DOIUrl":"https://doi.org/10.1109/CIST.2016.7805076","url":null,"abstract":"Traditional Information Retrieval (IR) models are based on bag-of-words paradigm, where relevance scores are computed based on exact matching of keywords. Although these models have already achieved good performance, it has been shown that most of dissatisfaction cases in relevance are due to term mismatch between queries and documents. In this paper, we introduce novel method to compute term frequency based on semantic similarities using distributed representations of words in a vector space (Word Embeddings). Our main goal is to allow distinct but semantically related terms to match each other and contribute to the relevance scores. Hence, Arabic documents are retrieved beyond the bag-of-words paradigm based on semantic similarities between word vectors. The results on Arabic standard TREC data sets show significant improvement over the baseline bag-of-words models.","PeriodicalId":196827,"journal":{"name":"2016 4th IEEE International Colloquium on Information Science and Technology (CiSt)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128868901","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 : 2016-10-01DOI: 10.1109/CIST.2016.7805052
Akachar Elyazid, B. Ouhbi, B. Frikh
In this paper, we studied some methods used in community detection in social networks. In the context of social networks, a community is a set of entities with a lot of interactions among them and little interaction with other sets outside. There are approaches related to static social networks, and others which focus on the dynamic social networks whose structure (actors and links) evolves over time. Static community detection approaches are able to find a division only if a graph is defined for a given time. However, many real graphs have the property to change and evolve over time. That is some nodes and links can appear or disappear during this process of evolution. In order to find communities in such networks we must take into account their different stages of evolution to provide coherent communities not just in a particular point in time, but all along their possible modifications over time. In this paper, we studied the static and dynamic approaches and we made a comparison between different algorithms using several real datasets.
{"title":"A comparative study of some algorithms for detecting communities in social networks","authors":"Akachar Elyazid, B. Ouhbi, B. Frikh","doi":"10.1109/CIST.2016.7805052","DOIUrl":"https://doi.org/10.1109/CIST.2016.7805052","url":null,"abstract":"In this paper, we studied some methods used in community detection in social networks. In the context of social networks, a community is a set of entities with a lot of interactions among them and little interaction with other sets outside. There are approaches related to static social networks, and others which focus on the dynamic social networks whose structure (actors and links) evolves over time. Static community detection approaches are able to find a division only if a graph is defined for a given time. However, many real graphs have the property to change and evolve over time. That is some nodes and links can appear or disappear during this process of evolution. In order to find communities in such networks we must take into account their different stages of evolution to provide coherent communities not just in a particular point in time, but all along their possible modifications over time. In this paper, we studied the static and dynamic approaches and we made a comparison between different algorithms using several real datasets.","PeriodicalId":196827,"journal":{"name":"2016 4th IEEE International Colloquium on Information Science and Technology (CiSt)","volume":"93 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115239076","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 : 2016-10-01DOI: 10.1109/CIST.2016.7805109
Meriem Hnida, Mohammed Khalidi, S. Bennani
In this paper, an evolutionary algorithm is applied to tackle Intelligent Curriculum Sequencing issue. The purpose is to align educational technology, for instance, curriculum sequence to: students' characteristics and subject-matter coherence. The algorithm considers both technical and pedagogical point of view. Results show that the proposed Evolutionary Computation Search Algorithm could find optimal learning sequences, under a set of constraints, within a reasonable amount of iterations.
{"title":"A novel approach for smart curriculum sequencing based on HSA evolutionary computation","authors":"Meriem Hnida, Mohammed Khalidi, S. Bennani","doi":"10.1109/CIST.2016.7805109","DOIUrl":"https://doi.org/10.1109/CIST.2016.7805109","url":null,"abstract":"In this paper, an evolutionary algorithm is applied to tackle Intelligent Curriculum Sequencing issue. The purpose is to align educational technology, for instance, curriculum sequence to: students' characteristics and subject-matter coherence. The algorithm considers both technical and pedagogical point of view. Results show that the proposed Evolutionary Computation Search Algorithm could find optimal learning sequences, under a set of constraints, within a reasonable amount of iterations.","PeriodicalId":196827,"journal":{"name":"2016 4th IEEE International Colloquium on Information Science and Technology (CiSt)","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122696939","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 : 2016-10-01DOI: 10.1109/CIST.2016.7805102
N. Zellal, A. Zaouia
The quality of data in data warehouse is very important for decision making. That's why we have been interested in studying the factors that influence this data quality. The first influencing factor is the quality of data at the source, and there are many other factors moderating this influence between data quality at the source and data quality at the data warehouse. This is what we have tried to gather in a research model in a previous work [1]. In order to validate this research model, we would like to conduct a survey and analyze the results using Structural Equation Modeling. And to prepare the survey, it is necessary to figure out the measured variables for each latent variable in the research model. So we aim, in this paper, to propose measurement items for each construct refereeing to the previous works and adapting them to our context.
{"title":"A measurement model for factors influencing data quality in data warehouse","authors":"N. Zellal, A. Zaouia","doi":"10.1109/CIST.2016.7805102","DOIUrl":"https://doi.org/10.1109/CIST.2016.7805102","url":null,"abstract":"The quality of data in data warehouse is very important for decision making. That's why we have been interested in studying the factors that influence this data quality. The first influencing factor is the quality of data at the source, and there are many other factors moderating this influence between data quality at the source and data quality at the data warehouse. This is what we have tried to gather in a research model in a previous work [1]. In order to validate this research model, we would like to conduct a survey and analyze the results using Structural Equation Modeling. And to prepare the survey, it is necessary to figure out the measured variables for each latent variable in the research model. So we aim, in this paper, to propose measurement items for each construct refereeing to the previous works and adapting them to our context.","PeriodicalId":196827,"journal":{"name":"2016 4th IEEE International Colloquium on Information Science and Technology (CiSt)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114500287","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 : 2016-10-01DOI: 10.1109/CIST.2016.7804852
Lotfi Najdi, Brahim Er-Raha
The identification of students' typologies plays interesting role in adapting educational strategies and improving academic performances. In this work, we show how unsupervised learning techniques can be applied to educational data for the extraction of typologies and profiles of graduate students based on educational outcomes in combination with the time to degree. We also describe a web-based tool for clustering student's data, based on R programming and shiny, in order to make the clustering analysis, more accessible for university decision maker. The clustering tool presented in this article will enhance the understanding of different learning characteristics of graduate students and could be used to adapt teaching approaches and strategies according to the identified student profiles.
{"title":"Implementing cluster analysis tool for the identification of students typologies","authors":"Lotfi Najdi, Brahim Er-Raha","doi":"10.1109/CIST.2016.7804852","DOIUrl":"https://doi.org/10.1109/CIST.2016.7804852","url":null,"abstract":"The identification of students' typologies plays interesting role in adapting educational strategies and improving academic performances. In this work, we show how unsupervised learning techniques can be applied to educational data for the extraction of typologies and profiles of graduate students based on educational outcomes in combination with the time to degree. We also describe a web-based tool for clustering student's data, based on R programming and shiny, in order to make the clustering analysis, more accessible for university decision maker. The clustering tool presented in this article will enhance the understanding of different learning characteristics of graduate students and could be used to adapt teaching approaches and strategies according to the identified student profiles.","PeriodicalId":196827,"journal":{"name":"2016 4th IEEE International Colloquium on Information Science and Technology (CiSt)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121996655","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 : 2016-10-01DOI: 10.1109/CIST.2016.7804966
M. F. Amr, K. Mansouri, A. Naji
The problem which companies want to face is the integration of the applications in an information system (IS) and the interconnection of the system itself with other information systems. The interoperability of IS constitutes, their ability to communicate and to inter-operate following a given model using a common language favoring the sharing of information for better performance and profitability. [1] In this article we reminder of the model that we have proposed for the interoperability of IS, it is a model of interconnection having for objective the synchronization of the business process of IS interconnected. This synchronization is based mainly on the transformation of business processes BPMN (Business Process Model Notation) of the different IS in Executable Languages Business Process Execution Language (BPEL) thanks to a language ATL (ATLAS Transformation Language) then bring together all the BPEL in a single global BPEL to basis of the rules of adaptation and an ontological database. Then and using the reverse engineering, to make an update, well adapted, Process BPMN trades of different IS. The key point of this architecture is the use of the ontological database. We present in this article a new hybrid model of ontological database based on two approaches of schema of ontological storage different that we were able to create from the result of analysis of a comparative study of database systems to ontological basis (BDBO). This analysis has helped us to adopt a hybrid model of BDBO who owns his own architecture and its storage model. This hybrid model that we are going to adopt for the implementation of the interoperability model of several information systems.
{"title":"Toward the development of a hybrid model of ontological database for the interoperability of several information systems","authors":"M. F. Amr, K. Mansouri, A. Naji","doi":"10.1109/CIST.2016.7804966","DOIUrl":"https://doi.org/10.1109/CIST.2016.7804966","url":null,"abstract":"The problem which companies want to face is the integration of the applications in an information system (IS) and the interconnection of the system itself with other information systems. The interoperability of IS constitutes, their ability to communicate and to inter-operate following a given model using a common language favoring the sharing of information for better performance and profitability. [1] In this article we reminder of the model that we have proposed for the interoperability of IS, it is a model of interconnection having for objective the synchronization of the business process of IS interconnected. This synchronization is based mainly on the transformation of business processes BPMN (Business Process Model Notation) of the different IS in Executable Languages Business Process Execution Language (BPEL) thanks to a language ATL (ATLAS Transformation Language) then bring together all the BPEL in a single global BPEL to basis of the rules of adaptation and an ontological database. Then and using the reverse engineering, to make an update, well adapted, Process BPMN trades of different IS. The key point of this architecture is the use of the ontological database. We present in this article a new hybrid model of ontological database based on two approaches of schema of ontological storage different that we were able to create from the result of analysis of a comparative study of database systems to ontological basis (BDBO). This analysis has helped us to adopt a hybrid model of BDBO who owns his own architecture and its storage model. This hybrid model that we are going to adopt for the implementation of the interoperability model of several information systems.","PeriodicalId":196827,"journal":{"name":"2016 4th IEEE International Colloquium on Information Science and Technology (CiSt)","volume":"2110 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129941585","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 : 2016-10-01DOI: 10.1109/CIST.2016.7804996
Chaimae El Hatri, J. Boumhidi
Traffic signal operations play an important role in the effective functioning of the urban area. However, due to the increasing number of vehicles and the high dynamic of the traffic network, conventional traffic signal timing methods does not result in an efficient control. One alternative is to let traffic signal controllers learn how to adjust the lights based on the traffic situation. In this paper, we propose a novel multi-objective traffic light control system that is based on an Intelligent Multi-Objective Particle Swarm Optimization (MOPSO) method. We take the average junction waiting time and the flow rate of vehicles on the congested road as two objectives. In the proposed method, we granted the ability of selecting appropriate MOPSO parameters to each agent of the swarm via a novel multi-objective Q-Learning approach. The simulation results demonstrate the efficiency of the proposed system.
{"title":"Q-learning based intelligent multi-objective particle swarm optimization of light control for traffic urban congestion management","authors":"Chaimae El Hatri, J. Boumhidi","doi":"10.1109/CIST.2016.7804996","DOIUrl":"https://doi.org/10.1109/CIST.2016.7804996","url":null,"abstract":"Traffic signal operations play an important role in the effective functioning of the urban area. However, due to the increasing number of vehicles and the high dynamic of the traffic network, conventional traffic signal timing methods does not result in an efficient control. One alternative is to let traffic signal controllers learn how to adjust the lights based on the traffic situation. In this paper, we propose a novel multi-objective traffic light control system that is based on an Intelligent Multi-Objective Particle Swarm Optimization (MOPSO) method. We take the average junction waiting time and the flow rate of vehicles on the congested road as two objectives. In the proposed method, we granted the ability of selecting appropriate MOPSO parameters to each agent of the swarm via a novel multi-objective Q-Learning approach. The simulation results demonstrate the efficiency of the proposed system.","PeriodicalId":196827,"journal":{"name":"2016 4th IEEE International Colloquium on Information Science and Technology (CiSt)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129080774","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 : 2016-10-01DOI: 10.1109/CIST.2016.7805037
Meryem Boufim, Hafid Barka
Due to the exponential growth of internet using, the habits of internet users have changed. This generate huge amount of data. It becomes important to explore this mine of knowledge and take advance on concurrent. In context of digital marketing, the user's data is the enterprise assets to personalize the content of websites and establish contact and communication with customers through internet channels. The more the enterprise has data and knows about its non-real customers the more it has the advantage to take advance and be leader. In this context inbound marking was recently born: to help marketers establishing customer-centric digital marketing strategy. On the other hand, Web mining is used to discover information through web data and construct knowledge about online customers. The web mining techniques seems to fit the best with the inbound marketing implementation. After an introduction to inbound marketing and web mining methods, this paper presents the application of web mining methods and techniques to implement an inbound marketing strategy.
{"title":"Converting strangers to clients using web mining techniques","authors":"Meryem Boufim, Hafid Barka","doi":"10.1109/CIST.2016.7805037","DOIUrl":"https://doi.org/10.1109/CIST.2016.7805037","url":null,"abstract":"Due to the exponential growth of internet using, the habits of internet users have changed. This generate huge amount of data. It becomes important to explore this mine of knowledge and take advance on concurrent. In context of digital marketing, the user's data is the enterprise assets to personalize the content of websites and establish contact and communication with customers through internet channels. The more the enterprise has data and knows about its non-real customers the more it has the advantage to take advance and be leader. In this context inbound marking was recently born: to help marketers establishing customer-centric digital marketing strategy. On the other hand, Web mining is used to discover information through web data and construct knowledge about online customers. The web mining techniques seems to fit the best with the inbound marketing implementation. After an introduction to inbound marketing and web mining methods, this paper presents the application of web mining methods and techniques to implement an inbound marketing strategy.","PeriodicalId":196827,"journal":{"name":"2016 4th IEEE International Colloquium on Information Science and Technology (CiSt)","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123668836","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 : 2016-10-01DOI: 10.1109/CIST.2016.7805038
Tiamaz Younes, Souissi Nissrine
Each company defines its goals to achieve an improvement approach for its business processes. However, identifying potentially achievable goals before the implementation of an improvement plan could prevent the company from wasting considerable time and budget on unreasonable goals that do not meet their expectations. Our meta-analysis provides, in this sense, a preliminary study of expected objectives through the implementation of the Lean approach in different fields and by an analysis of several cases published in the literature.
{"title":"Meta-analysis of lean case studies for the business process improvement","authors":"Tiamaz Younes, Souissi Nissrine","doi":"10.1109/CIST.2016.7805038","DOIUrl":"https://doi.org/10.1109/CIST.2016.7805038","url":null,"abstract":"Each company defines its goals to achieve an improvement approach for its business processes. However, identifying potentially achievable goals before the implementation of an improvement plan could prevent the company from wasting considerable time and budget on unreasonable goals that do not meet their expectations. Our meta-analysis provides, in this sense, a preliminary study of expected objectives through the implementation of the Lean approach in different fields and by an analysis of several cases published in the literature.","PeriodicalId":196827,"journal":{"name":"2016 4th IEEE International Colloquium on Information Science and Technology (CiSt)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121154088","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 : 2016-10-01DOI: 10.1109/CIST.2016.7805043
Amine Agharghor, M. E. Riffi, Faycal Chebihi
Since 1930s, traveling salesman problem is still one of the most studied problems in optimization. It started to be used as a benchmark for the new optimization methods that solves the combinatorial optimization problem NP-hard. This paper proposes an assessment of a memetic Hunting Search algorithm that uses a 2-Opt local search for solving the traveling salesman problem. Hunting Search is an evolutionary algorithm inspired by the method of group hunting of predatory animals. To show the quality of the memetic algorithm, it has been checked on a set of ten benchmark TSPLib instances and it outperforms the results obtained with previous Hunting Search algorithm.
{"title":"A memetic hunting search algorithm for the traveling salesman problem","authors":"Amine Agharghor, M. E. Riffi, Faycal Chebihi","doi":"10.1109/CIST.2016.7805043","DOIUrl":"https://doi.org/10.1109/CIST.2016.7805043","url":null,"abstract":"Since 1930s, traveling salesman problem is still one of the most studied problems in optimization. It started to be used as a benchmark for the new optimization methods that solves the combinatorial optimization problem NP-hard. This paper proposes an assessment of a memetic Hunting Search algorithm that uses a 2-Opt local search for solving the traveling salesman problem. Hunting Search is an evolutionary algorithm inspired by the method of group hunting of predatory animals. To show the quality of the memetic algorithm, it has been checked on a set of ten benchmark TSPLib instances and it outperforms the results obtained with previous Hunting Search algorithm.","PeriodicalId":196827,"journal":{"name":"2016 4th IEEE International Colloquium on Information Science and Technology (CiSt)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121321113","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：