C. Navarrete, M. D. L. Cruz, E. Anguiano, A. Ortega, J. Rojas
This paper compares two different approaches, followed by our research group, to efficiently run NEPs on parallel platforms, as general and transparent as possible. The vague results of jNEP (our multithreaded Java simulator for multicore desktop computers) suggests the use of massively parallel platforms (clusters of computers). The good results obtained show the scalability and viability of this last approach.
{"title":"Parallel Simulation of NEPs on Clusters","authors":"C. Navarrete, M. D. L. Cruz, E. Anguiano, A. Ortega, J. Rojas","doi":"10.1109/WI-IAT.2011.131","DOIUrl":"https://doi.org/10.1109/WI-IAT.2011.131","url":null,"abstract":"This paper compares two different approaches, followed by our research group, to efficiently run NEPs on parallel platforms, as general and transparent as possible. The vague results of jNEP (our multithreaded Java simulator for multicore desktop computers) suggests the use of massively parallel platforms (clusters of computers). The good results obtained show the scalability and viability of this last approach.","PeriodicalId":128421,"journal":{"name":"2011 IEEE/WIC/ACM International Conferences on Web Intelligence and Intelligent Agent Technology","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122469536","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 paper, we introduce a mobile device enabled multi-agent system that aims at more effective information integration for large-scale plant management. In our approach, different agents are in charge of individual plant units, plant activities (e.g., scheduling), and human-computer interface. Agents residing in mobile devices enable operators to access critical plant information, process data, and make decisions in a remote and around-the-clock fashion. The effectiveness of the system has been demonstrated with an implementation to manage a virtual plan.
{"title":"Agile Plant Management Using Agents and Mobile Devices: Enhancing Collaboration and Information Integration in Large-Scale","authors":"Benito Mendoza, Peng Xu, Qiangguo Ren, Li Bai","doi":"10.1109/WI-IAT.2011.257","DOIUrl":"https://doi.org/10.1109/WI-IAT.2011.257","url":null,"abstract":"In this paper, we introduce a mobile device enabled multi-agent system that aims at more effective information integration for large-scale plant management. In our approach, different agents are in charge of individual plant units, plant activities (e.g., scheduling), and human-computer interface. Agents residing in mobile devices enable operators to access critical plant information, process data, and make decisions in a remote and around-the-clock fashion. The effectiveness of the system has been demonstrated with an implementation to manage a virtual plan.","PeriodicalId":128421,"journal":{"name":"2011 IEEE/WIC/ACM International Conferences on Web Intelligence and Intelligent Agent Technology","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125534724","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}
Recent work has provided tantalizing hints that small amounts of cooperation may actually hurt a group's performance rather than help it. In this paper, we take a systematic look at the value of cooperation. Using a simple cooperative task where agents can act effectively individually but where high levels of cooperation will intuitively lead to better behavior, we investigated when and how cooperation helped overall performance. We systematically varied properties of the environment, e.g., the amount of uncertainty, and how much of what sort of cooperation the agents would perform, e.g., information sharing, resource allocation etc. Our experimental results show that, even if communication were free, and while typically coordination helps the team, under some circumstances, the team may be better off not coordinating at all than coordinating a little bit. We show that the level of uncertainty the agents face and their initial understandings of the environment determine whether a small amount of coordination is useful.
{"title":"Towards an Understanding of the Value of Cooperation in Uncertain World","authors":"Yonghong Wang, K. Sycara, P. Scerri","doi":"10.1109/WI-IAT.2011.244","DOIUrl":"https://doi.org/10.1109/WI-IAT.2011.244","url":null,"abstract":"Recent work has provided tantalizing hints that small amounts of cooperation may actually hurt a group's performance rather than help it. In this paper, we take a systematic look at the value of cooperation. Using a simple cooperative task where agents can act effectively individually but where high levels of cooperation will intuitively lead to better behavior, we investigated when and how cooperation helped overall performance. We systematically varied properties of the environment, e.g., the amount of uncertainty, and how much of what sort of cooperation the agents would perform, e.g., information sharing, resource allocation etc. Our experimental results show that, even if communication were free, and while typically coordination helps the team, under some circumstances, the team may be better off not coordinating at all than coordinating a little bit. We show that the level of uncertainty the agents face and their initial understandings of the environment determine whether a small amount of coordination is useful.","PeriodicalId":128421,"journal":{"name":"2011 IEEE/WIC/ACM International Conferences on Web Intelligence and Intelligent Agent Technology","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130511960","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 proposes a collective intelligence strategy for emerging semantic data. It presents a combination of social web practices with semantic web technologies to enrich existing web resources with semantic data. The paper introduces a social semantic tagging approach called Semdrops. Semdrops defines a conceptual model which is an extension of the Gruber's tag model where the tag concept is extended to semantic tag. Semdrops is implemented as a Firefox add-on tool that turns the web browser into a collaborative semantic data editor. To validate Semdrops's approach, we conducted an evaluation and usability studies and compared the results with automatic generation methods of semantic data such as DBpedia. The studies demonstrated that Semdrops is an effective and complementary approach to produce adequate semantic data on the Web.
{"title":"Semdrops: A Social Semantic Tagging Approach for Emerging Semantic Data","authors":"Diego Torres, A. Díaz, H. Skaf-Molli, P. Molli","doi":"10.1109/WI-IAT.2011.51","DOIUrl":"https://doi.org/10.1109/WI-IAT.2011.51","url":null,"abstract":"This paper proposes a collective intelligence strategy for emerging semantic data. It presents a combination of social web practices with semantic web technologies to enrich existing web resources with semantic data. The paper introduces a social semantic tagging approach called Semdrops. Semdrops defines a conceptual model which is an extension of the Gruber's tag model where the tag concept is extended to semantic tag. Semdrops is implemented as a Firefox add-on tool that turns the web browser into a collaborative semantic data editor. To validate Semdrops's approach, we conducted an evaluation and usability studies and compared the results with automatic generation methods of semantic data such as DBpedia. The studies demonstrated that Semdrops is an effective and complementary approach to produce adequate semantic data on the Web.","PeriodicalId":128421,"journal":{"name":"2011 IEEE/WIC/ACM International Conferences on Web Intelligence and Intelligent Agent Technology","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123374739","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 volume of email that help-desks receive every day is very high and often queries are repeated. Any kind of automation in processing of emails requires good understanding of the emails. In the current work we propose a schema for tagging author composed sentences in help-desk emails by the intent of the author. We have created a corpus taking email data from two help-desks and annotated them at sentence level. We have achieved significant accuracies in learning to automatically tag the sentences using n-gram features and some hand-picked lexical features. At every stage right from choice of schema to choice of features, we have tried to be domain independent or keep domain related information as a separate component. Automation of Tagging of Discourse Segments (TODS) in email, we propose is a significant step towards finding the discourse parse of emails.
{"title":"Novel Approach for Tagging of Discourse Segments in Help-Desk E-Mails","authors":"Suman Yelati, R. Sangal","doi":"10.1109/WI-IAT.2011.71","DOIUrl":"https://doi.org/10.1109/WI-IAT.2011.71","url":null,"abstract":"The volume of email that help-desks receive every day is very high and often queries are repeated. Any kind of automation in processing of emails requires good understanding of the emails. In the current work we propose a schema for tagging author composed sentences in help-desk emails by the intent of the author. We have created a corpus taking email data from two help-desks and annotated them at sentence level. We have achieved significant accuracies in learning to automatically tag the sentences using n-gram features and some hand-picked lexical features. At every stage right from choice of schema to choice of features, we have tried to be domain independent or keep domain related information as a separate component. Automation of Tagging of Discourse Segments (TODS) in email, we propose is a significant step towards finding the discourse parse of emails.","PeriodicalId":128421,"journal":{"name":"2011 IEEE/WIC/ACM International Conferences on Web Intelligence and Intelligent Agent Technology","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116462807","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}
Computational approaches to language acquisition typically reduce the problem to one of either learning syntax, or learning a lexicon -- isolated tasks. These problem simplifications have led to disconnected solutions and algorithms with no expectation of grounded inputs. We propose a language learning framework for bootstrapping the acquisition of a grammar and learning lexical semantics. We describe the framework and provide an instantiation using context-free grammars and a lambda-calculus semantic representation.
{"title":"An Architecture for Bootstrapping Lexical Semantics and Grammatical Structures","authors":"T. Armstrong, T. Oates","doi":"10.1109/WI-IAT.2011.188","DOIUrl":"https://doi.org/10.1109/WI-IAT.2011.188","url":null,"abstract":"Computational approaches to language acquisition typically reduce the problem to one of either learning syntax, or learning a lexicon -- isolated tasks. These problem simplifications have led to disconnected solutions and algorithms with no expectation of grounded inputs. We propose a language learning framework for bootstrapping the acquisition of a grammar and learning lexical semantics. We describe the framework and provide an instantiation using context-free grammars and a lambda-calculus semantic representation.","PeriodicalId":128421,"journal":{"name":"2011 IEEE/WIC/ACM International Conferences on Web Intelligence and Intelligent Agent Technology","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130288065","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}
Vincent Verdot, B. Christophe, V. Toubiana, Mathieu Beauvais
Information overload is a real challenge becoming more and more critical regarding the quite limited human capabilities. 107 trillion emails sent in 2010, that is more than we could even imagine so how to help the user to take advantage from this? That is the question we address in Scribee, a multidisciplinary Bell Labs initiative which ocuses on the very pragmatic case of email service. From a knowledge and conversation-based approach we try to improve the email technologies in the context of information overload. Scribee itself has not yet provided its conclusions but the experiment offers interesting statistics on email: half of messages' content is useless, most conversations ends within a business week, emails are not equal, etc. This analysis based on real email conversations provides valuable information about this "old" communication service and could help improving it.
{"title":"Scribee Experimentation - Early Statistics on Email Conversations","authors":"Vincent Verdot, B. Christophe, V. Toubiana, Mathieu Beauvais","doi":"10.1109/WI-IAT.2011.48","DOIUrl":"https://doi.org/10.1109/WI-IAT.2011.48","url":null,"abstract":"Information overload is a real challenge becoming more and more critical regarding the quite limited human capabilities. 107 trillion emails sent in 2010, that is more than we could even imagine so how to help the user to take advantage from this? That is the question we address in Scribee, a multidisciplinary Bell Labs initiative which ocuses on the very pragmatic case of email service. From a knowledge and conversation-based approach we try to improve the email technologies in the context of information overload. Scribee itself has not yet provided its conclusions but the experiment offers interesting statistics on email: half of messages' content is useless, most conversations ends within a business week, emails are not equal, etc. This analysis based on real email conversations provides valuable information about this \"old\" communication service and could help improving it.","PeriodicalId":128421,"journal":{"name":"2011 IEEE/WIC/ACM International Conferences on Web Intelligence and Intelligent Agent Technology","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130296211","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 article, we tackle the problem of coalition formation in hedonic games. In such context, agents have their own preferences over coalitions which they may join and could be endowed with different behavioral strategies in order to search for appropriate partners and better assess the proposals of the other agents. However when agents are self interested and free to decide which strategies they use, the system cannot expect them to form a coalition structure with specific properties. Our paper addresses such issue and studies the constraints that could be enforced on self-interested agents, in order to form coalitions which guarantee significant solution concepts. In this article, we focus on the Pareto optimality of the formed coalition structure. We develop our coalition formation model and propose a theoretical analysis of this model.
{"title":"Constraining Self-Interested Agents to Guarantee Pareto Optimality in Multiagent Coalition Formation Problem","authors":"T. Génin, S. Aknine","doi":"10.1109/WI-IAT.2011.179","DOIUrl":"https://doi.org/10.1109/WI-IAT.2011.179","url":null,"abstract":"In this article, we tackle the problem of coalition formation in hedonic games. In such context, agents have their own preferences over coalitions which they may join and could be endowed with different behavioral strategies in order to search for appropriate partners and better assess the proposals of the other agents. However when agents are self interested and free to decide which strategies they use, the system cannot expect them to form a coalition structure with specific properties. Our paper addresses such issue and studies the constraints that could be enforced on self-interested agents, in order to form coalitions which guarantee significant solution concepts. In this article, we focus on the Pareto optimality of the formed coalition structure. We develop our coalition formation model and propose a theoretical analysis of this model.","PeriodicalId":128421,"journal":{"name":"2011 IEEE/WIC/ACM International Conferences on Web Intelligence and Intelligent Agent Technology","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134031509","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}
Natural language processing (NLP) is becoming much more robust and applicable in realistic applications. One area in which NLP has still not been fully exploited is information retrieval (IR). In particular we are interested in search over intranets and other local Web sites. We see dialogue-driven search which is based on a largely automated knowledge extraction process as one of the next big steps. Instead of replying with a set of documents for a user query the system would allow the user to navigate through the extracted knowledge base by making use of a simple dialogue manager. Here we support this idea with a first task-based evaluation that we conducted on a university intranet. We automatically extracted entities like person names, organizations and locations as well as relations between entities and added visual graphs to the search results whenever a user query could be mapped into this knowledge base. We found that users are willing to interact and use those visual interfaces. We also found that users preferred such a system that guides a user through the result set over a baseline approach. The results represent an important first step towards full NLP-driven intranet search.
{"title":"Combining the Best of Two Worlds: NLP and IR for Intranet Search","authors":"Suma Adindla, Udo Kruschwitz","doi":"10.1109/WI-IAT.2011.187","DOIUrl":"https://doi.org/10.1109/WI-IAT.2011.187","url":null,"abstract":"Natural language processing (NLP) is becoming much more robust and applicable in realistic applications. One area in which NLP has still not been fully exploited is information retrieval (IR). In particular we are interested in search over intranets and other local Web sites. We see dialogue-driven search which is based on a largely automated knowledge extraction process as one of the next big steps. Instead of replying with a set of documents for a user query the system would allow the user to navigate through the extracted knowledge base by making use of a simple dialogue manager. Here we support this idea with a first task-based evaluation that we conducted on a university intranet. We automatically extracted entities like person names, organizations and locations as well as relations between entities and added visual graphs to the search results whenever a user query could be mapped into this knowledge base. We found that users are willing to interact and use those visual interfaces. We also found that users preferred such a system that guides a user through the result set over a baseline approach. The results represent an important first step towards full NLP-driven intranet search.","PeriodicalId":128421,"journal":{"name":"2011 IEEE/WIC/ACM International Conferences on Web Intelligence and Intelligent Agent Technology","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133730922","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}
Businesses of all sizes have now realized that the Web is an invaluable resource for competitive intelligence, and consequently business decision making. But many have trouble collecting targeted & useful information, and are often further overwhelmed by the time required for analysis & monitoring. On another hand, text mining techniques have become widely used for information analysis in the scientific community in general, and are now ubiquitous in most Web Intelligence fields. With the availability of services such as Google Prediction API, or mature open source software such as GATE, RapidMiner or NLTK, one can expect a much wider adoption of text mining and associated machine learning techniques by expert developers. But how can these techniques benefit to the daily life of a wider business audience? As competitive intelligence is often focused on products, people, customers and competitors, there is an added value for systems providing analytics on these entities, whose recognition is fundamental to text mining and semantic analysis, and consequently is still under active scientific investigation. In this talk we will tour some of the specific requirements and options for building an efficient Web based competitive intelligence system with named entity analytics. We will see how some savvy simplifications can help to overcome common issues such as Web scale and Web content noise, and finally deliver acceptable usability and value for non-specialists, business users.
{"title":"Extracting Named Entities at Web Scale for Competitive Intelligence","authors":"François Pouilloux","doi":"10.1109/WI-IAT.2011.284","DOIUrl":"https://doi.org/10.1109/WI-IAT.2011.284","url":null,"abstract":"Businesses of all sizes have now realized that the Web is an invaluable resource for competitive intelligence, and consequently business decision making. But many have trouble collecting targeted & useful information, and are often further overwhelmed by the time required for analysis & monitoring. On another hand, text mining techniques have become widely used for information analysis in the scientific community in general, and are now ubiquitous in most Web Intelligence fields. With the availability of services such as Google Prediction API, or mature open source software such as GATE, RapidMiner or NLTK, one can expect a much wider adoption of text mining and associated machine learning techniques by expert developers. But how can these techniques benefit to the daily life of a wider business audience? As competitive intelligence is often focused on products, people, customers and competitors, there is an added value for systems providing analytics on these entities, whose recognition is fundamental to text mining and semantic analysis, and consequently is still under active scientific investigation. In this talk we will tour some of the specific requirements and options for building an efficient Web based competitive intelligence system with named entity analytics. We will see how some savvy simplifications can help to overcome common issues such as Web scale and Web content noise, and finally deliver acceptable usability and value for non-specialists, business users.","PeriodicalId":128421,"journal":{"name":"2011 IEEE/WIC/ACM International Conferences on Web Intelligence and Intelligent Agent Technology","volume":"261 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132256198","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：