Pub Date : 2017-11-01DOI: 10.1109/ISKE.2017.8258831
Haihong Dai, B. Hu, Qian Cui, Z. Zou
With the rapid development of smart cities and the advance of city safety and defense demands, the question that how to accurately discover and retrieve the data that users request from VideoGIS data faces a series of bottleneck problems. VideoGIS data retrieval is one of the important ways to solve above problems. In order to accelerate the rate of feature matching and improve the efficiency of the video retrieval, a new method of VideoGIS data retrieval based on multi-feature fusion is proposed in this paper. The method firstly use video frame difference based on Euclidean distance to extract the key frames under the spatial and temporal sampling of the video. On the basis of this, the global features (e.g. color, shape, texture) are fused by different weighted coefficients, then the feature vector, as the video multi-feature fusion representation, can be constructed by fusing the global features and local features. Based on the multi-feature fusion, correlation between video features is made full use. Compared with the method of single feature and two-feature fusion, the experimental results indicate that the proposed retrieval method has better retrieval effect.
{"title":"VideoGIS data retrieval based on multi-feature fusion","authors":"Haihong Dai, B. Hu, Qian Cui, Z. Zou","doi":"10.1109/ISKE.2017.8258831","DOIUrl":"https://doi.org/10.1109/ISKE.2017.8258831","url":null,"abstract":"With the rapid development of smart cities and the advance of city safety and defense demands, the question that how to accurately discover and retrieve the data that users request from VideoGIS data faces a series of bottleneck problems. VideoGIS data retrieval is one of the important ways to solve above problems. In order to accelerate the rate of feature matching and improve the efficiency of the video retrieval, a new method of VideoGIS data retrieval based on multi-feature fusion is proposed in this paper. The method firstly use video frame difference based on Euclidean distance to extract the key frames under the spatial and temporal sampling of the video. On the basis of this, the global features (e.g. color, shape, texture) are fused by different weighted coefficients, then the feature vector, as the video multi-feature fusion representation, can be constructed by fusing the global features and local features. Based on the multi-feature fusion, correlation between video features is made full use. Compared with the method of single feature and two-feature fusion, the experimental results indicate that the proposed retrieval method has better retrieval effect.","PeriodicalId":208009,"journal":{"name":"2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117152275","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 : 2017-11-01DOI: 10.1109/ISKE.2017.8258741
Shuwei Chen, Yang Xu, Yan Jiang, Jun Liu, Xingxing He
The synergized dynamic contradiction separation based automated deduction theory has provided a novel logic based automated deduction reasoning framework, which extends the static binary resolution inference rule to a dynamic multiple contradiction separation based automated deduction mechanism. This novel contradiction separation based automated deduction mechanism is characterized as a dynamic, multi-clauses involving, synergized, goal-oriented and robust automated reasoning framework. In order to further improve the efficiency and feasibility of this novel automated deduction mechanism, this paper proposes some strategies for clause or literal selection during the automated deduction process, which consider mainly the synergized effect of multi-clauses during the deduction process. Some examples are put forward to illustrate the feasibility of these proposed strategies.
{"title":"Some synergized clause selection strategies for contradiction separation based automated deduction","authors":"Shuwei Chen, Yang Xu, Yan Jiang, Jun Liu, Xingxing He","doi":"10.1109/ISKE.2017.8258741","DOIUrl":"https://doi.org/10.1109/ISKE.2017.8258741","url":null,"abstract":"The synergized dynamic contradiction separation based automated deduction theory has provided a novel logic based automated deduction reasoning framework, which extends the static binary resolution inference rule to a dynamic multiple contradiction separation based automated deduction mechanism. This novel contradiction separation based automated deduction mechanism is characterized as a dynamic, multi-clauses involving, synergized, goal-oriented and robust automated reasoning framework. In order to further improve the efficiency and feasibility of this novel automated deduction mechanism, this paper proposes some strategies for clause or literal selection during the automated deduction process, which consider mainly the synergized effect of multi-clauses during the deduction process. Some examples are put forward to illustrate the feasibility of these proposed strategies.","PeriodicalId":208009,"journal":{"name":"2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122144373","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 : 2017-11-01DOI: 10.1109/ISKE.2017.8258764
P. Flores-Vidal, D. Gómez, J. Montero, G. Villarino
Edge detection problems try to identify those pixels that represent the boundaries of the objects in an image. The process for getting a solution is usually organized in several steps, producing at the end a set of pixels that could be edges (candidates to be edges). These pixels are then classified based on some local evaluation method, taking into account the measurements obtained in each pixel. In this paper, we propose a global evaluation method based on the idea of edge list to produce a solution. In particular, we propose an algorithm divided in four steps: in first place we build the edge list (that we have called segments); in second place we extract the characteristics associated to each segment (length, intensity, location,…); in the third step we learn which are the characteristics that make a segment good enough to be a boundary; finally, in the fourth place, we apply the classification task. In this work we have built the ground truth of edge list necessary for the supervised classification. Finally we test the effectiveness of this algorithm against other classical approaches.
{"title":"Classifying segments in edge detection problems","authors":"P. Flores-Vidal, D. Gómez, J. Montero, G. Villarino","doi":"10.1109/ISKE.2017.8258764","DOIUrl":"https://doi.org/10.1109/ISKE.2017.8258764","url":null,"abstract":"Edge detection problems try to identify those pixels that represent the boundaries of the objects in an image. The process for getting a solution is usually organized in several steps, producing at the end a set of pixels that could be edges (candidates to be edges). These pixels are then classified based on some local evaluation method, taking into account the measurements obtained in each pixel. In this paper, we propose a global evaluation method based on the idea of edge list to produce a solution. In particular, we propose an algorithm divided in four steps: in first place we build the edge list (that we have called segments); in second place we extract the characteristics associated to each segment (length, intensity, location,…); in the third step we learn which are the characteristics that make a segment good enough to be a boundary; finally, in the fourth place, we apply the classification task. In this work we have built the ground truth of edge list necessary for the supervised classification. Finally we test the effectiveness of this algorithm against other classical approaches.","PeriodicalId":208009,"journal":{"name":"2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122159310","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 : 2017-11-01DOI: 10.1109/ISKE.2017.8258799
Hongbin Liu, Ickjai Lee
Transportation mode classification is a key task in trajectory data mining. It adds human behaviour semantics to raw trajectories for trip recommendation, traffic management and transport planning. Previous approaches require heavy pre-processing and feature extraction processes in order to build a classifier, which is complicated and time-consuming. Recurrent neural network has demonstrated its capacity in sequence modelling tasks ranging from machine translation, speech recognition to image captioning. In this paper, we propose a trajectory transportation mode classification framework that is based on an end-to-end bidirectional LSTM classifier. The proposed classification process does not require any feature extraction process, but automatically learns features from trajectories, and use them for classification. We further improve this framework by feeding the time interval as an external feature by embedding. Our experiments on real GPS datasets demonstrate that our approach outperforms existing methods with regard to AUC.
{"title":"End-to-end trajectory transportation mode classification using Bi-LSTM recurrent neural network","authors":"Hongbin Liu, Ickjai Lee","doi":"10.1109/ISKE.2017.8258799","DOIUrl":"https://doi.org/10.1109/ISKE.2017.8258799","url":null,"abstract":"Transportation mode classification is a key task in trajectory data mining. It adds human behaviour semantics to raw trajectories for trip recommendation, traffic management and transport planning. Previous approaches require heavy pre-processing and feature extraction processes in order to build a classifier, which is complicated and time-consuming. Recurrent neural network has demonstrated its capacity in sequence modelling tasks ranging from machine translation, speech recognition to image captioning. In this paper, we propose a trajectory transportation mode classification framework that is based on an end-to-end bidirectional LSTM classifier. The proposed classification process does not require any feature extraction process, but automatically learns features from trajectories, and use them for classification. We further improve this framework by feeding the time interval as an external feature by embedding. Our experiments on real GPS datasets demonstrate that our approach outperforms existing methods with regard to AUC.","PeriodicalId":208009,"journal":{"name":"2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123969304","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 : 2017-11-01DOI: 10.1109/ISKE.2017.8258738
Qifeng Zhou, Ruifeng Yuan, Tao Li
Analyzing and understanding disaster-related situation updates from a large number of disaster-related documents plays an important role in disaster management and has attracted a lot of research attention. Recently several methods have been developed to generate textual storylines from disaster-related documents to help people understand the overall trend and evolution of a disaster event as well as how the disaster affects different areas. These methods are able to help people improve their situation awareness by generating informative summarizes to present the global pictures of disaster events. However, these methods suffer from several limitations including text representation, representative document selection, and summary generation that may affect the quality of the summarized results. To address these limitations, in this paper, we propose an improved two-layer storyline generating framework which generates a global storyline of the disaster events in the first layer, and provides condensed information about specific regions affected by the disaster in the second layer. The proposed framework utilizes the word embedding for text similarity measurement, considers both uniqueness and relevance for representative document selection, and uses Maximal Marginal Relevance to generate summaries from each local document set. The experimental results on four typhoons related events demonstrate the efficacy of our proposed framework on capturing the status information and understanding the situation from a large of documents.
{"title":"An improved textual storyline generating framework for disaster information management","authors":"Qifeng Zhou, Ruifeng Yuan, Tao Li","doi":"10.1109/ISKE.2017.8258738","DOIUrl":"https://doi.org/10.1109/ISKE.2017.8258738","url":null,"abstract":"Analyzing and understanding disaster-related situation updates from a large number of disaster-related documents plays an important role in disaster management and has attracted a lot of research attention. Recently several methods have been developed to generate textual storylines from disaster-related documents to help people understand the overall trend and evolution of a disaster event as well as how the disaster affects different areas. These methods are able to help people improve their situation awareness by generating informative summarizes to present the global pictures of disaster events. However, these methods suffer from several limitations including text representation, representative document selection, and summary generation that may affect the quality of the summarized results. To address these limitations, in this paper, we propose an improved two-layer storyline generating framework which generates a global storyline of the disaster events in the first layer, and provides condensed information about specific regions affected by the disaster in the second layer. The proposed framework utilizes the word embedding for text similarity measurement, considers both uniqueness and relevance for representative document selection, and uses Maximal Marginal Relevance to generate summaries from each local document set. The experimental results on four typhoons related events demonstrate the efficacy of our proposed framework on capturing the status information and understanding the situation from a large of documents.","PeriodicalId":208009,"journal":{"name":"2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE)","volume":"564 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116286507","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 : 2017-11-01DOI: 10.1109/ISKE.2017.8258788
Yan Ma, Zining Cao, Yang Liu
At present, there are various logics used to describe properties of continue-time systems, such as TCTL (timed computation tree logic) and RTCTL (rewards timed computation tree logic), and of continue-time stochastic systems, such as CSL (continuous stochastic logic) and PTCTL (probabilistic timed computation tree logic). When assigned on the systems, these logics are limited to returning only true or false. In order to establish the correctness and the performance of a system one needs a unified framework for expressing the system, and a unified logic for expressing the desired specification. In this word, we propose a new formal specification TCTML (timed computation tree measurement language) to check the correctness and evaluate the performance of a system, which operates on real values to reason over probabilistic continue-time systems. Retaining the conventional semantics of temporal logics, it extends weak until and until with arithmetic operations on real-valued quantities. Moreover, the algorithms of CTML model checking PTAs (probabilistic timed automata) is given and a concrete example is used to demonstrate that it is feasible and effective.
{"title":"A specification with performance evaluation for probabilistic timed automata","authors":"Yan Ma, Zining Cao, Yang Liu","doi":"10.1109/ISKE.2017.8258788","DOIUrl":"https://doi.org/10.1109/ISKE.2017.8258788","url":null,"abstract":"At present, there are various logics used to describe properties of continue-time systems, such as TCTL (timed computation tree logic) and RTCTL (rewards timed computation tree logic), and of continue-time stochastic systems, such as CSL (continuous stochastic logic) and PTCTL (probabilistic timed computation tree logic). When assigned on the systems, these logics are limited to returning only true or false. In order to establish the correctness and the performance of a system one needs a unified framework for expressing the system, and a unified logic for expressing the desired specification. In this word, we propose a new formal specification TCTML (timed computation tree measurement language) to check the correctness and evaluate the performance of a system, which operates on real values to reason over probabilistic continue-time systems. Retaining the conventional semantics of temporal logics, it extends weak until and until with arithmetic operations on real-valued quantities. Moreover, the algorithms of CTML model checking PTAs (probabilistic timed automata) is given and a concrete example is used to demonstrate that it is feasible and effective.","PeriodicalId":208009,"journal":{"name":"2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114759879","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 : 2017-11-01DOI: 10.1109/ISKE.2017.8258830
Xianzhong Long, Lei Chen, Qun Li
As one kind of popular application in computer vision, image clustering has attracted many attentions. Some machine learning algorithms have been widely employed, such as K-Means, Non-negative Matrix Factorization (NMF), Graph regularized Non-negative Matrix Factorization (GNMF) and Locally Consistent Concept Factorization (LCCF). These methods possess respective strength and weakness. The common problem existing in these clustering algorithms is that they only use one kind of feature. However, different kinds of features complement each other and can be used to improve performance results. In this paper, in order to make use of the complementarity between different features, we propose an image representation method based on multi-features. Clustering results on several benchmark image data sets show that the proposed scheme outperforms some classical methods.
{"title":"Image representation based on multi-features","authors":"Xianzhong Long, Lei Chen, Qun Li","doi":"10.1109/ISKE.2017.8258830","DOIUrl":"https://doi.org/10.1109/ISKE.2017.8258830","url":null,"abstract":"As one kind of popular application in computer vision, image clustering has attracted many attentions. Some machine learning algorithms have been widely employed, such as K-Means, Non-negative Matrix Factorization (NMF), Graph regularized Non-negative Matrix Factorization (GNMF) and Locally Consistent Concept Factorization (LCCF). These methods possess respective strength and weakness. The common problem existing in these clustering algorithms is that they only use one kind of feature. However, different kinds of features complement each other and can be used to improve performance results. In this paper, in order to make use of the complementarity between different features, we propose an image representation method based on multi-features. Clustering results on several benchmark image data sets show that the proposed scheme outperforms some classical methods.","PeriodicalId":208009,"journal":{"name":"2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114770375","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 : 2017-11-01DOI: 10.1109/ISKE.2017.8258767
A. Masood, Adel Al-Jumaily
Development of automated diagnosis systems using machine learning and expert knowledge based data analysis requires effective automated learning models. However, models based on limited expert labeled training data can wrongly affect the results of diagnosis due to insufficient training knowledge acquired. On the other hand, getting more relevant analytical details from all the data used for training is an aspect that can enhance the efficiency of learning algorithms. This paper proposes a semi-advised training and classification algorithm that has the capability to effectively use limited labeled data along with abundant unlabeled data. It demonstrates the capability to use unlabeled data for training the algorithm by obtaining sufficient amount of information through incorporating an advised and/or partially supervised methodology. For comparative analysis, dermatological and histopathalogical images of skin cancer are used as experimental datasets. The proposed algorithm provided very impressive diagnosis outputs for both type of datasets in comparison to several other famous algorithms that are usually used in literature for classification.
{"title":"Semi advised learning and classification algorithm for partially labeled skin cancer data analysis","authors":"A. Masood, Adel Al-Jumaily","doi":"10.1109/ISKE.2017.8258767","DOIUrl":"https://doi.org/10.1109/ISKE.2017.8258767","url":null,"abstract":"Development of automated diagnosis systems using machine learning and expert knowledge based data analysis requires effective automated learning models. However, models based on limited expert labeled training data can wrongly affect the results of diagnosis due to insufficient training knowledge acquired. On the other hand, getting more relevant analytical details from all the data used for training is an aspect that can enhance the efficiency of learning algorithms. This paper proposes a semi-advised training and classification algorithm that has the capability to effectively use limited labeled data along with abundant unlabeled data. It demonstrates the capability to use unlabeled data for training the algorithm by obtaining sufficient amount of information through incorporating an advised and/or partially supervised methodology. For comparative analysis, dermatological and histopathalogical images of skin cancer are used as experimental datasets. The proposed algorithm provided very impressive diagnosis outputs for both type of datasets in comparison to several other famous algorithms that are usually used in literature for classification.","PeriodicalId":208009,"journal":{"name":"2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126345336","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 : 2017-11-01DOI: 10.1109/ISKE.2017.8258757
Mushtaq Ahmad, Qingchun Chen, M. Najam-ul-Islam, M. Iqbal, S. Hussain
A plethora of routing protocols are available for effective and efficient data dissemination in MANETs. But due to broadcast nature of wireless communications, these protocols are prone to diverse security attacks. In this work, the Secure Optimized Link State Routing (SOLSR) protocol is proposed to fill the breach for security attacks while finding path(s) from source to destination by including message authentication code (MAC) value that is appended to message as a signature, followed by encryption and Global secret key in both HELLO messages and Topology control (TC) messages. Timestamp mechanism is used to avoid replay attacks because authentication and encryption cannot prevent replay attacks. This paper also presents the NS-2 simulation results to assess the SOLSR protocol and the achieved performance comparison between SOLSR and OLSR over different network scenarios. It is shown that the proposed SOLSR outperforms the OLSR in terms of the achieved security, data overhead, packet loss and delay.
{"title":"On the secure optimized link state routing (SOLSR) protocol for MANETs","authors":"Mushtaq Ahmad, Qingchun Chen, M. Najam-ul-Islam, M. Iqbal, S. Hussain","doi":"10.1109/ISKE.2017.8258757","DOIUrl":"https://doi.org/10.1109/ISKE.2017.8258757","url":null,"abstract":"A plethora of routing protocols are available for effective and efficient data dissemination in MANETs. But due to broadcast nature of wireless communications, these protocols are prone to diverse security attacks. In this work, the Secure Optimized Link State Routing (SOLSR) protocol is proposed to fill the breach for security attacks while finding path(s) from source to destination by including message authentication code (MAC) value that is appended to message as a signature, followed by encryption and Global secret key in both HELLO messages and Topology control (TC) messages. Timestamp mechanism is used to avoid replay attacks because authentication and encryption cannot prevent replay attacks. This paper also presents the NS-2 simulation results to assess the SOLSR protocol and the achieved performance comparison between SOLSR and OLSR over different network scenarios. It is shown that the proposed SOLSR outperforms the OLSR in terms of the achieved security, data overhead, packet loss and delay.","PeriodicalId":208009,"journal":{"name":"2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125604769","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 : 2017-11-01DOI: 10.1109/ISKE.2017.8258791
Hamdi Ellouzi, Mounir Ben Ayed, Hela Ltifi
Assigning a set of intelligent agents in each phase of the Knowledge Discovery from Data (KDD) process improves communication and cooperation between the different KDD steps in order to generate relevant results knowledge for decision-making tasks. We aim in particular to simulate the behavior of a KDD system in which specific intelligent agents interact with each other and with their environment. Such environment is characterized by its dynamic, visual and real-time aspects. The proposed approach is applied to develop a prototype for the fight against nosocomial infections in the hospital Intensive Care Units. This paper ends with the evaluation of the prototype usability.
{"title":"Modeling of distributed visual knowledge discovery from data process","authors":"Hamdi Ellouzi, Mounir Ben Ayed, Hela Ltifi","doi":"10.1109/ISKE.2017.8258791","DOIUrl":"https://doi.org/10.1109/ISKE.2017.8258791","url":null,"abstract":"Assigning a set of intelligent agents in each phase of the Knowledge Discovery from Data (KDD) process improves communication and cooperation between the different KDD steps in order to generate relevant results knowledge for decision-making tasks. We aim in particular to simulate the behavior of a KDD system in which specific intelligent agents interact with each other and with their environment. Such environment is characterized by its dynamic, visual and real-time aspects. The proposed approach is applied to develop a prototype for the fight against nosocomial infections in the hospital Intensive Care Units. This paper ends with the evaluation of the prototype usability.","PeriodicalId":208009,"journal":{"name":"2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124364853","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：