This work presents IBMAP, an approach for robust learning of Markov network structures from data, together with IBMAP-HC, an efficient instantiation of the approach. Existing Score-Based (SB) and Independence-Based (IB) approaches must make concessions either on robustness or efficiency. IBMAP-HC improves robustness efficiently through an IB-SB hybrid approach based on the probabilistic Maximum-A-Posteriori (MAP) technique, and the IB-score, a tractable expression for computing posterior probabilities of Markov network structures. Performance is first tested against IB and SB competitors on synthetic datasets. Against IB competitors (GSMN algorithm and a version of the HHC algorithm adapted here for Markov networks discovery), IBMAP-HC showed reductions in edges Hamming distance with same order running times. Against SB competitors, both IBMAP-HC and our adaptation of HHC produced comparable Hamming distances, but with running times orders of magnitude faster. We also evaluated IBMAP-HC in a realistic, challenging test-bed: EDAs, evolutionary algorithms for optimization that estimate a distribution on each generation. Using IBMAP-HC to estimate distributions, EDAs converged to the optimum faster in all benchmark functions considered, reducing required fitness evaluations by up to 80%.
{"title":"Independence-Based MAP for Markov Networks Structure Discovery","authors":"F. Bromberg, F. Schlüter, A. Edera","doi":"10.1109/ICTAI.2011.81","DOIUrl":"https://doi.org/10.1109/ICTAI.2011.81","url":null,"abstract":"This work presents IBMAP, an approach for robust learning of Markov network structures from data, together with IBMAP-HC, an efficient instantiation of the approach. Existing Score-Based (SB) and Independence-Based (IB) approaches must make concessions either on robustness or efficiency. IBMAP-HC improves robustness efficiently through an IB-SB hybrid approach based on the probabilistic Maximum-A-Posteriori (MAP) technique, and the IB-score, a tractable expression for computing posterior probabilities of Markov network structures. Performance is first tested against IB and SB competitors on synthetic datasets. Against IB competitors (GSMN algorithm and a version of the HHC algorithm adapted here for Markov networks discovery), IBMAP-HC showed reductions in edges Hamming distance with same order running times. Against SB competitors, both IBMAP-HC and our adaptation of HHC produced comparable Hamming distances, but with running times orders of magnitude faster. We also evaluated IBMAP-HC in a realistic, challenging test-bed: EDAs, evolutionary algorithms for optimization that estimate a distribution on each generation. Using IBMAP-HC to estimate distributions, EDAs converged to the optimum faster in all benchmark functions considered, reducing required fitness evaluations by up to 80%.","PeriodicalId":332661,"journal":{"name":"2011 IEEE 23rd International Conference on Tools with Artificial Intelligence","volume":"102 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117117966","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}
In multi-agent recommender systems, agents are able to generate recommendations according to the preferences of the customer. However, in some domains, specific knowledge is required in order to compose a recommendation and this knowledge may be not available for the agent. In these cases, agents need to communicate with other agents in the community searching for the specific information to complete the recommendation. This paper presents a multi-agent recommender system based on trust and expert agents. It aims at improving the quality of the information exchanged among agents because communication will occur primarily with trusted sources in the hope to decrease the communication load. Also, agents become experts in specific types of recommendation. The approach was validate in the tourism domain by means of recommendations of travel packages and experiments were performed to illustrate the impact of using trust assignment in the quality of the recommendations generated by expert agents. Results corroborate the intuition that expert agents that use a trust mechanism are able to increase the quality of recommendation provided.
{"title":"Enhancing the Quality of Recommendations through Expert and Trusted Agents","authors":"Fabiana Lorenzi, Mara Abel, S. Loh, André Peres","doi":"10.1109/ICTAI.2011.56","DOIUrl":"https://doi.org/10.1109/ICTAI.2011.56","url":null,"abstract":"In multi-agent recommender systems, agents are able to generate recommendations according to the preferences of the customer. However, in some domains, specific knowledge is required in order to compose a recommendation and this knowledge may be not available for the agent. In these cases, agents need to communicate with other agents in the community searching for the specific information to complete the recommendation. This paper presents a multi-agent recommender system based on trust and expert agents. It aims at improving the quality of the information exchanged among agents because communication will occur primarily with trusted sources in the hope to decrease the communication load. Also, agents become experts in specific types of recommendation. The approach was validate in the tourism domain by means of recommendations of travel packages and experiments were performed to illustrate the impact of using trust assignment in the quality of the recommendations generated by expert agents. Results corroborate the intuition that expert agents that use a trust mechanism are able to increase the quality of recommendation provided.","PeriodicalId":332661,"journal":{"name":"2011 IEEE 23rd International Conference on Tools with Artificial Intelligence","volume":"122 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115512028","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}
This paper addresses the problem of failure diagnosis in component-based discrete event systems. In this paper we propose a method to obtain the set of components when dealing with diagnosis in large complex discrete event systems. In the new method, before disassembling the system into components, we need to identify whether insert communication events into the system or not. When analyzing the diagnosability, we treat the system containing communication events as a distributed discrete event system. Otherwise we treat the system as a decentralized discrete event system. For the components which are not diagnosable, we propose a method to reconstruct them by utilizing some other components sharing the same communication events with them. This algorithm provides more accurate information of the diagnosability of the system.
{"title":"Disassembling and Reconstructing Algorithms for Discrete Event Systems","authors":"Xuena Geng, D. Ouyang, Jinsong Guo, Xiaoyu Wang","doi":"10.1109/ICTAI.2011.101","DOIUrl":"https://doi.org/10.1109/ICTAI.2011.101","url":null,"abstract":"This paper addresses the problem of failure diagnosis in component-based discrete event systems. In this paper we propose a method to obtain the set of components when dealing with diagnosis in large complex discrete event systems. In the new method, before disassembling the system into components, we need to identify whether insert communication events into the system or not. When analyzing the diagnosability, we treat the system containing communication events as a distributed discrete event system. Otherwise we treat the system as a decentralized discrete event system. For the components which are not diagnosable, we propose a method to reconstruct them by utilizing some other components sharing the same communication events with them. This algorithm provides more accurate information of the diagnosability of the system.","PeriodicalId":332661,"journal":{"name":"2011 IEEE 23rd International Conference on Tools with Artificial Intelligence","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114646367","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}
The multiobjective route planning problem in dynamic multi-hop ridesharing is considered to be NP-complete. Evolutionary computation has received a growing interest in solving the hard multiobjective optimization problems. In this study we investigate the behavior of different variants of the ant colony based approach for solving the multiobjective route planning problem and compare the performance of the different variants with the performance of a genetic algorithm recommended for solving the problem. Experimentation results indicate that the ant colony approach encounters poor performance in its native form and competes the genetic approach in some of its variants when combined with local search.
{"title":"Ant Colony vs. Genetic Multiobjective Route Planning in Dynamic Multi-hop Ridesharing","authors":"Wesam Herbawi, M. Weber","doi":"10.1109/ICTAI.2011.50","DOIUrl":"https://doi.org/10.1109/ICTAI.2011.50","url":null,"abstract":"The multiobjective route planning problem in dynamic multi-hop ridesharing is considered to be NP-complete. Evolutionary computation has received a growing interest in solving the hard multiobjective optimization problems. In this study we investigate the behavior of different variants of the ant colony based approach for solving the multiobjective route planning problem and compare the performance of the different variants with the performance of a genetic algorithm recommended for solving the problem. Experimentation results indicate that the ant colony approach encounters poor performance in its native form and competes the genetic approach in some of its variants when combined with local search.","PeriodicalId":332661,"journal":{"name":"2011 IEEE 23rd International Conference on Tools with Artificial Intelligence","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125931524","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}
R. Karimi, C. Freudenthaler, A. Nanopoulos, L. Schmidt-Thieme
Recommender systems help web users to address information overload. However their performance depends on the number of provided ratings by users. This problem is amplified for a new user because he/she has not provided any rating. To address this problem, active learning methods have been proposed to acquire those ratings from users, that will help most in determining their interests. The optimal active learning selects a query that directly optimizes the expected error for the test data. This approach is applicable for prediction models in which this question can be answered in closed-form given the distribution of test data is known. Unfortunately, there are many tasks and models for which the optimal selection cannot efficiently be found in closed-form. Therefore, most of the active learning methods optimize different, non-optimal criteria, such as uncertainty. Nevertheless, in this paper we exploit the characteristics of matrix factorization, which leads to a closed-form solution and by being inspired from existing optimal active learning for the regression task, develop a method that approximates the optimal solution for recommender systems. Our results demonstrate that the proposed method improves the prediction accuracy of MF.
{"title":"Towards Optimal Active Learning for Matrix Factorization in Recommender Systems","authors":"R. Karimi, C. Freudenthaler, A. Nanopoulos, L. Schmidt-Thieme","doi":"10.1109/ICTAI.2011.182","DOIUrl":"https://doi.org/10.1109/ICTAI.2011.182","url":null,"abstract":"Recommender systems help web users to address information overload. However their performance depends on the number of provided ratings by users. This problem is amplified for a new user because he/she has not provided any rating. To address this problem, active learning methods have been proposed to acquire those ratings from users, that will help most in determining their interests. The optimal active learning selects a query that directly optimizes the expected error for the test data. This approach is applicable for prediction models in which this question can be answered in closed-form given the distribution of test data is known. Unfortunately, there are many tasks and models for which the optimal selection cannot efficiently be found in closed-form. Therefore, most of the active learning methods optimize different, non-optimal criteria, such as uncertainty. Nevertheless, in this paper we exploit the characteristics of matrix factorization, which leads to a closed-form solution and by being inspired from existing optimal active learning for the regression task, develop a method that approximates the optimal solution for recommender systems. Our results demonstrate that the proposed method improves the prediction accuracy of MF.","PeriodicalId":332661,"journal":{"name":"2011 IEEE 23rd International Conference on Tools with Artificial Intelligence","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123683776","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}
Galois lattices' (GLs) definition is defined for a binary table (called context). Therefore, in the presence of continuous data, a discretization step is needed. Discretization is classically performed before the lattice construction in a global way. However, local discretization is reported to give better classification rates than global discretization when used jointly with other symbolic classification methods such as decision trees (DTs). We present a new algorithm performing local discretization for GLs using the lattice properties. Our local discretization algorithm is applied iteratively to particular nodes (called concepts) of the GL. Experiments are performed to assess the efficiency and the effectiveness of the proposed algorithm compared to global discretization.
{"title":"Local Discretization of Numerical Data for Galois Lattices","authors":"Nathalie Girard, K. Bertet, M. Visani","doi":"10.1109/ICTAI.2011.148","DOIUrl":"https://doi.org/10.1109/ICTAI.2011.148","url":null,"abstract":"Galois lattices' (GLs) definition is defined for a binary table (called context). Therefore, in the presence of continuous data, a discretization step is needed. Discretization is classically performed before the lattice construction in a global way. However, local discretization is reported to give better classification rates than global discretization when used jointly with other symbolic classification methods such as decision trees (DTs). We present a new algorithm performing local discretization for GLs using the lattice properties. Our local discretization algorithm is applied iteratively to particular nodes (called concepts) of the GL. Experiments are performed to assess the efficiency and the effectiveness of the proposed algorithm compared to global discretization.","PeriodicalId":332661,"journal":{"name":"2011 IEEE 23rd International Conference on Tools with Artificial Intelligence","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125385402","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}
Mohamad El Falou, M. Bouzid, A. Mouaddib, Thierry Vidal
The establishment of a self-healing agent has received much interest in multiple domains such as : Web services, production supply chain, transport systems, etc. This agent has a set of actions. Its role is to respond to user request with a plan of composed actions, to on-line diagnose the status of the plan execution and to automatically repair the plan when a fault is detected during the plan's execution. To this end, three main areas are studied and modeled for the establishment of such an agent : composition, diagnosis and repair.
{"title":"A Self Healing Action Composition Agent","authors":"Mohamad El Falou, M. Bouzid, A. Mouaddib, Thierry Vidal","doi":"10.1109/ICTAI.2011.90","DOIUrl":"https://doi.org/10.1109/ICTAI.2011.90","url":null,"abstract":"The establishment of a self-healing agent has received much interest in multiple domains such as : Web services, production supply chain, transport systems, etc. This agent has a set of actions. Its role is to respond to user request with a plan of composed actions, to on-line diagnose the status of the plan execution and to automatically repair the plan when a fault is detected during the plan's execution. To this end, three main areas are studied and modeled for the establishment of such an agent : composition, diagnosis and repair.","PeriodicalId":332661,"journal":{"name":"2011 IEEE 23rd International Conference on Tools with Artificial Intelligence","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122014720","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}
This paper introduces a clustering algorithm based on batch Self-Organizing Maps to partition objects taking into account their relational descriptions given by multiple dissimilarity matrices. The presented approach provides a partition of the objects and a prototype for each cluster, moreover the method is able to learn relevance weights for each dissimilarity matrix by optimizing an adequacy criterion that measures the fit between clusters and the respective prototypes. These relevance weights change at each iteration and are different from one cluster to another.
{"title":"Adaptive Batch SOM for Multiple Dissimilarity Data Tables","authors":"Anderson Dantas, F. D. Carvalho","doi":"10.1109/ICTAI.2011.92","DOIUrl":"https://doi.org/10.1109/ICTAI.2011.92","url":null,"abstract":"This paper introduces a clustering algorithm based on batch Self-Organizing Maps to partition objects taking into account their relational descriptions given by multiple dissimilarity matrices. The presented approach provides a partition of the objects and a prototype for each cluster, moreover the method is able to learn relevance weights for each dissimilarity matrix by optimizing an adequacy criterion that measures the fit between clusters and the respective prototypes. These relevance weights change at each iteration and are different from one cluster to another.","PeriodicalId":332661,"journal":{"name":"2011 IEEE 23rd International Conference on Tools with Artificial Intelligence","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129247636","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}
Camera surveillance systems have proved useful for public safety. The main disadvantage is that since the camera views are monitored in a remote control room, it is often difficult for security officers to reach the crime-scene in time. During this time, the assailant(s) have likely caused sufficient damage and threaten many lives. To overcome this problem, this paper proposes to take a standard surveillance system, augment it with simple weaponry for the purpose of disabling potential assailants, and use mathematical models to develop decision criteria for selecting the safest and most effective weapon in a given situation. The feasibility of the proposed system is examined using simulation results which also validate the utility of the proposed decision models.
{"title":"Toward a Remote-Controlled Weapon-Equipped Camera Surveillance System","authors":"Robert Bisewski, P. Atrey","doi":"10.1109/ICTAI.2011.185","DOIUrl":"https://doi.org/10.1109/ICTAI.2011.185","url":null,"abstract":"Camera surveillance systems have proved useful for public safety. The main disadvantage is that since the camera views are monitored in a remote control room, it is often difficult for security officers to reach the crime-scene in time. During this time, the assailant(s) have likely caused sufficient damage and threaten many lives. To overcome this problem, this paper proposes to take a standard surveillance system, augment it with simple weaponry for the purpose of disabling potential assailants, and use mathematical models to develop decision criteria for selecting the safest and most effective weapon in a given situation. The feasibility of the proposed system is examined using simulation results which also validate the utility of the proposed decision models.","PeriodicalId":332661,"journal":{"name":"2011 IEEE 23rd International Conference on Tools with Artificial Intelligence","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129899665","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}
In this work we adopt a previously introduced meta-learning classification method for semi-supervised learning problems. In our previous work we illustrated that the method is successful when applied in a supervised classification problem. In our current work the results demonstrate that following refinements made to the method it can be successfully applied to semi-supervised classification cases.
{"title":"Classification by Clusters Analysis - An Ensemble Technique in a Semi-supervised Classification","authors":"A. Jurek, Y. Bi, Shengli Wu, C. Nugent","doi":"10.1109/ICTAI.2011.137","DOIUrl":"https://doi.org/10.1109/ICTAI.2011.137","url":null,"abstract":"In this work we adopt a previously introduced meta-learning classification method for semi-supervised learning problems. In our previous work we illustrated that the method is successful when applied in a supervised classification problem. In our current work the results demonstrate that following refinements made to the method it can be successfully applied to semi-supervised classification cases.","PeriodicalId":332661,"journal":{"name":"2011 IEEE 23rd International Conference on Tools with Artificial Intelligence","volume":"494 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127031051","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：