This work presents a new perspective on characterizing the similarity between elements of a database or, more generally, nodes of a weighted, undirected graph. It is based on a Markov-chain model of random walk through the database. The suggested quantities, representing dissimilarities (or similarities) between any two elements, have the nice property of decreasing (increasing) when the number of paths connecting those elements increases and when the "length" of any path decreases. The model is evaluated on a collaborative recommendation task where suggestions are made about which movies people should watch based upon what they watched in the past. The model, which nicely fits into the so-called "statistical relational learning" framework as well as the "link analysis" paradigm, could also be used to compute document or word similarities, and, more generally could be applied to other database or Web mining tasks.
{"title":"A novel way of computing similarities between nodes of a graph, with application to collaborative recommendation","authors":"François Fouss, A. Pirotte, M. Saerens","doi":"10.1109/WI.2005.9","DOIUrl":"https://doi.org/10.1109/WI.2005.9","url":null,"abstract":"This work presents a new perspective on characterizing the similarity between elements of a database or, more generally, nodes of a weighted, undirected graph. It is based on a Markov-chain model of random walk through the database. The suggested quantities, representing dissimilarities (or similarities) between any two elements, have the nice property of decreasing (increasing) when the number of paths connecting those elements increases and when the \"length\" of any path decreases. The model is evaluated on a collaborative recommendation task where suggestions are made about which movies people should watch based upon what they watched in the past. The model, which nicely fits into the so-called \"statistical relational learning\" framework as well as the \"link analysis\" paradigm, could also be used to compute document or word similarities, and, more generally could be applied to other database or Web mining tasks.","PeriodicalId":213856,"journal":{"name":"The 2005 IEEE/WIC/ACM International Conference on Web Intelligence (WI'05)","volume":"160 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116637151","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}
Schema matching is a complex process focusing on matching between concepts describing the data in heterogeneous data sources. There is a shift from manual schema matching, done by human experts, to automatic matching, using various heuristics (schema matchers). In this work, we consider the problem of linearly combining the results of a set of schema matchers. We propose the use of machine learning algorithms to learn the optimal weight assignments, given a set of schema matchers. We also suggest the use of genetic algorithms to improve the process efficiency.
{"title":"Measuring the relative performance of schema matchers","authors":"S. Berkovsky, Yaniv Eytani, A. Gal","doi":"10.1109/WI.2005.94","DOIUrl":"https://doi.org/10.1109/WI.2005.94","url":null,"abstract":"Schema matching is a complex process focusing on matching between concepts describing the data in heterogeneous data sources. There is a shift from manual schema matching, done by human experts, to automatic matching, using various heuristics (schema matchers). In this work, we consider the problem of linearly combining the results of a set of schema matchers. We propose the use of machine learning algorithms to learn the optimal weight assignments, given a set of schema matchers. We also suggest the use of genetic algorithms to improve the process efficiency.","PeriodicalId":213856,"journal":{"name":"The 2005 IEEE/WIC/ACM International Conference on Web Intelligence (WI'05)","volume":"58 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125710745","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}
Ioan Marius Bilasco, J. Gensel, M. Villanova-Oliver
This paper addresses the adaptation of dynamic and synchronized multimedia presentations built by querying XML compatible data sources. We provide WIS designers with facilities for describing presentations whose content is not known at design time in terms of quantity, but only after the execution of queries. Our approach relies on the definition of a template. A template consists of a model that aims at automatically adapting the multimedia content of a presentation to both the user's profile and the characteristics of her/his access device. We show here how a template is built and how adaptations of the presentation are performed when the quantity of information and/or the material capabilities of the access devices (e.g. display size), do not match the template's spatiotemporal specifications.
{"title":"STAMP: adaptable templates for synchronized multimedia presentations","authors":"Ioan Marius Bilasco, J. Gensel, M. Villanova-Oliver","doi":"10.1109/WI.2005.137","DOIUrl":"https://doi.org/10.1109/WI.2005.137","url":null,"abstract":"This paper addresses the adaptation of dynamic and synchronized multimedia presentations built by querying XML compatible data sources. We provide WIS designers with facilities for describing presentations whose content is not known at design time in terms of quantity, but only after the execution of queries. Our approach relies on the definition of a template. A template consists of a model that aims at automatically adapting the multimedia content of a presentation to both the user's profile and the characteristics of her/his access device. We show here how a template is built and how adaptations of the presentation are performed when the quantity of information and/or the material capabilities of the access devices (e.g. display size), do not match the template's spatiotemporal specifications.","PeriodicalId":213856,"journal":{"name":"The 2005 IEEE/WIC/ACM International Conference on Web Intelligence (WI'05)","volume":"430 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126096449","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}
Online, single-pass mining Web click streams poses some interesting computational issues, such as unbounded length of streaming data, possibly very fast arrival rate and just one scan over previously arrived click-sequencer In this paper, we propose a new, single-pass algorithm, called DSM-TKP (data stream mining for top-k path traversal patterns), for mining top-k path traversal patterns, where k is the desired number of path traversal patterns to be mined. An effective summary data structure called TKP-forest (top-k path forest) is used to maintain the essential information about the top-k path traversal patterns of the click-stream so far. Experimental studies show that DSM-TKP algorithm uses stable memory usage and makes only one pass over the streaming data.
{"title":"DSM-TKP: mining top-k path traversal patterns over Web click-streams","authors":"Hua-Fu Li, Suh-Yin Lee, M. Shan","doi":"10.1109/WI.2005.56","DOIUrl":"https://doi.org/10.1109/WI.2005.56","url":null,"abstract":"Online, single-pass mining Web click streams poses some interesting computational issues, such as unbounded length of streaming data, possibly very fast arrival rate and just one scan over previously arrived click-sequencer In this paper, we propose a new, single-pass algorithm, called DSM-TKP (data stream mining for top-k path traversal patterns), for mining top-k path traversal patterns, where k is the desired number of path traversal patterns to be mined. An effective summary data structure called TKP-forest (top-k path forest) is used to maintain the essential information about the top-k path traversal patterns of the click-stream so far. Experimental studies show that DSM-TKP algorithm uses stable memory usage and makes only one pass over the streaming data.","PeriodicalId":213856,"journal":{"name":"The 2005 IEEE/WIC/ACM International Conference on Web Intelligence (WI'05)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123977658","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 describe the navigational behavior of the students of an e-learning virtual environment, in order to determine whether such navigational patterns are related to the academic performance achieved by the students or not, and which behaviors can be identified as more successful. As an example, a subset of students taking a degree in computer science in a completely virtual online university is selected as the matter of study. Three levels of analysis are described: a session level, where students perform a few actions in a single session logged to the virtual campus; a course level, where all single sessions are joined to form a course navigational pattern; and a lifelong learning level, where students enroll in several subjects each academic semester. A simple experiment is outlined for the course level to demonstrate the possibilities of such analysis in a virtual e-learning environment. This experiment shows that the information collected in this level is useful for understanding user behavior and the relationship with his or her academic achievements, and that some intuitive ideas about the relevance of specific user actions or particularities can be also better explained.
{"title":"User navigational behavior in e-learning virtual environments","authors":"J. M. Carbó, Enric Mor, J. Minguillón","doi":"10.1109/WI.2005.155","DOIUrl":"https://doi.org/10.1109/WI.2005.155","url":null,"abstract":"In this paper, we describe the navigational behavior of the students of an e-learning virtual environment, in order to determine whether such navigational patterns are related to the academic performance achieved by the students or not, and which behaviors can be identified as more successful. As an example, a subset of students taking a degree in computer science in a completely virtual online university is selected as the matter of study. Three levels of analysis are described: a session level, where students perform a few actions in a single session logged to the virtual campus; a course level, where all single sessions are joined to form a course navigational pattern; and a lifelong learning level, where students enroll in several subjects each academic semester. A simple experiment is outlined for the course level to demonstrate the possibilities of such analysis in a virtual e-learning environment. This experiment shows that the information collected in this level is useful for understanding user behavior and the relationship with his or her academic achievements, and that some intuitive ideas about the relevance of specific user actions or particularities can be also better explained.","PeriodicalId":213856,"journal":{"name":"The 2005 IEEE/WIC/ACM International Conference on Web Intelligence (WI'05)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124166420","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}
We propose a collaborative environment for semantic-enabled mobile devices (e.g. PDAs, cell phones, laptops) in peer to peer scenarios. Within the environment, resource discovery is performed exploiting technologies and techniques for knowledge representation developed for the semantic Web, which have been adapted to cope with the highly flexible structure of ad-hoc networks in ubiquitous computing. The approach exploits the standard Bluetooth stack, using the original UUID payload, to carry semantically annotated data. The environment is motivated and presented in a museum case study.
{"title":"Semantic based collaborative P2P in ubiquitous computing","authors":"M. Ruta, T. D. Noia, E. Sciascio, F. Donini","doi":"10.1109/WI.2005.130","DOIUrl":"https://doi.org/10.1109/WI.2005.130","url":null,"abstract":"We propose a collaborative environment for semantic-enabled mobile devices (e.g. PDAs, cell phones, laptops) in peer to peer scenarios. Within the environment, resource discovery is performed exploiting technologies and techniques for knowledge representation developed for the semantic Web, which have been adapted to cope with the highly flexible structure of ad-hoc networks in ubiquitous computing. The approach exploits the standard Bluetooth stack, using the original UUID payload, to carry semantically annotated data. The environment is motivated and presented in a museum case study.","PeriodicalId":213856,"journal":{"name":"The 2005 IEEE/WIC/ACM International Conference on Web Intelligence (WI'05)","volume":"2012 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127380192","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}
These days, many companies are executing their business aims based on decentralized cooperation of software components which work on various systems over a network. However, messages and processes between systems are designed individually in each operations division. Therefore, the system development for adjusting interfaces is expensive, so the companies cannot introduce their services in a dynamic business environment. To resolve such problems, we propose interface modeling technique and message mapping technique which model the relationship between the message formats and semantics on the formats by using Web Ontology Language (OWL) and execute the mapping between the message formats by using semantics. We developed the user interactive message mapping tool and evaluated our proposed methods based on the interface specifications of real network management systems.
{"title":"Proposal and verification of flexible interface mapping technique for automatic system cooperation based on semantics","authors":"M. Nakatsuji, Y. Miyoshi, Tatsuyuki Kimura","doi":"10.1109/WI.2005.120","DOIUrl":"https://doi.org/10.1109/WI.2005.120","url":null,"abstract":"These days, many companies are executing their business aims based on decentralized cooperation of software components which work on various systems over a network. However, messages and processes between systems are designed individually in each operations division. Therefore, the system development for adjusting interfaces is expensive, so the companies cannot introduce their services in a dynamic business environment. To resolve such problems, we propose interface modeling technique and message mapping technique which model the relationship between the message formats and semantics on the formats by using Web Ontology Language (OWL) and execute the mapping between the message formats by using semantics. We developed the user interactive message mapping tool and evaluated our proposed methods based on the interface specifications of real network management systems.","PeriodicalId":213856,"journal":{"name":"The 2005 IEEE/WIC/ACM International Conference on Web Intelligence (WI'05)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129850570","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 presents a system for automatically detecting and filtering unsolicited electronic messages. The underlying filtering method is based on email origin and content. A heuristic knowledge base formed by spam words is extracted from labelled emails by a finite state automata. The processing of three parts of every email by a single Bayesian filter and the integration of the every part classification allows to achieve a maximum performance goal. The system is dynamic and interactive and evolves from the evolution of spam by incremental machine learning.
{"title":"An interactive hybrid system for identifying and filtering unsolicited email","authors":"M. D. D. Castillo, J. I. Serrano","doi":"10.1109/WI.2005.31","DOIUrl":"https://doi.org/10.1109/WI.2005.31","url":null,"abstract":"This paper presents a system for automatically detecting and filtering unsolicited electronic messages. The underlying filtering method is based on email origin and content. A heuristic knowledge base formed by spam words is extracted from labelled emails by a finite state automata. The processing of three parts of every email by a single Bayesian filter and the integration of the every part classification allows to achieve a maximum performance goal. The system is dynamic and interactive and evolves from the evolution of spam by incremental machine learning.","PeriodicalId":213856,"journal":{"name":"The 2005 IEEE/WIC/ACM International Conference on Web Intelligence (WI'05)","volume":"2016 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128974582","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, a Web document classification scheme, progressive analysis scheme (PAS) is proposed to efficiently and effectively classify HTML Web documents. When an author writes a Web document, HTML tags are used to visually emphasize the texts related to main concepts. The design of PAS is to catch the authoring convention in terms of the contributions of nested HTML tags to document classification. During the learning phase, PAS provides an enhanced tag sequence model to resolve the sample lacking problem in learning the classification contributions of HTML tag sequences. While in classification phase, PAS decomposes a Web document into regions based on the DOM tag-tree, and analyzes the regions in the descending order of their classification contributions. PAS also provides a mechanism called emphasis degree adjustment to defer the processing of noisy region during classification. The simulation results shows that PAS has better performance than full-text (e.g. SVM) and sequential classifier.
{"title":"Progressive analysis scheme for Web document classification","authors":"Li-Chun Sung, Chin-Hwa Kuo, M. Chen, Yeali S. Sun","doi":"10.1109/WI.2005.119","DOIUrl":"https://doi.org/10.1109/WI.2005.119","url":null,"abstract":"In this paper, a Web document classification scheme, progressive analysis scheme (PAS) is proposed to efficiently and effectively classify HTML Web documents. When an author writes a Web document, HTML tags are used to visually emphasize the texts related to main concepts. The design of PAS is to catch the authoring convention in terms of the contributions of nested HTML tags to document classification. During the learning phase, PAS provides an enhanced tag sequence model to resolve the sample lacking problem in learning the classification contributions of HTML tag sequences. While in classification phase, PAS decomposes a Web document into regions based on the DOM tag-tree, and analyzes the regions in the descending order of their classification contributions. PAS also provides a mechanism called emphasis degree adjustment to defer the processing of noisy region during classification. The simulation results shows that PAS has better performance than full-text (e.g. SVM) and sequential classifier.","PeriodicalId":213856,"journal":{"name":"The 2005 IEEE/WIC/ACM International Conference on Web Intelligence (WI'05)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128017200","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}
Matthias Nickles, Ruth Cobos Pérez, Gerhard Weiss, Tina Froehner
In open environments like the Web, and open multiagent and Peer2Peer systems, consent among the autonomous, self-interested knowledge sources and users very often cannot be established, and the estimation of trustability and truthfulness of knowledge sources may not be possible. Moreover, competing viewpoints and their communicative contexts even provide valuable meta-knowledge about the intentions of the participants and their social relationships. As a foundational approach to semantically heterogeneous knowledge perspectives, we introduce a formal framework for the computational representation and integration of multi-source knowledge, which makes explicit heterogeneous viewpoints, and conflicting opinions and their social contexts, and allows for the rating, generalization and optional fusion of knowledge by social choice.
{"title":"Multi-source knowledge bases and ontologies with multiple individual and social viewpoints","authors":"Matthias Nickles, Ruth Cobos Pérez, Gerhard Weiss, Tina Froehner","doi":"10.1109/WI.2005.104","DOIUrl":"https://doi.org/10.1109/WI.2005.104","url":null,"abstract":"In open environments like the Web, and open multiagent and Peer2Peer systems, consent among the autonomous, self-interested knowledge sources and users very often cannot be established, and the estimation of trustability and truthfulness of knowledge sources may not be possible. Moreover, competing viewpoints and their communicative contexts even provide valuable meta-knowledge about the intentions of the participants and their social relationships. As a foundational approach to semantically heterogeneous knowledge perspectives, we introduce a formal framework for the computational representation and integration of multi-source knowledge, which makes explicit heterogeneous viewpoints, and conflicting opinions and their social contexts, and allows for the rating, generalization and optional fusion of knowledge by social choice.","PeriodicalId":213856,"journal":{"name":"The 2005 IEEE/WIC/ACM International Conference on Web Intelligence (WI'05)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117215250","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：