Map generalization is used to derive maps for secondary scales and/or specific goals. This operation greatly benefits spatial decision support systems as it can provide a global and simplified representation of a phenomenon discarding irrelevant information. The recent popularity of OLAP systems for various application domains has generated much interest for the development of spatial OLAP (SOLAP) models that integrate spatial data in data warehouse and OLAP systems. Although powerful under some respect, current SOLAP models cannot support map generalization capabilities. In this paper, we present a conceptual multidimensional model integrating map generalization. The model extends SOLAP spatial hierarchies introducing multi-association relationships, and supports imprecise measures.
{"title":"Enriching Spatial OLAP with Map Generalization: a Conceptual Multidimensional Model","authors":"S. Bimonte, J. Gensel, M. Bertolotto","doi":"10.1109/ICDMW.2008.80","DOIUrl":"https://doi.org/10.1109/ICDMW.2008.80","url":null,"abstract":"Map generalization is used to derive maps for secondary scales and/or specific goals. This operation greatly benefits spatial decision support systems as it can provide a global and simplified representation of a phenomenon discarding irrelevant information. The recent popularity of OLAP systems for various application domains has generated much interest for the development of spatial OLAP (SOLAP) models that integrate spatial data in data warehouse and OLAP systems. Although powerful under some respect, current SOLAP models cannot support map generalization capabilities. In this paper, we present a conceptual multidimensional model integrating map generalization. The model extends SOLAP spatial hierarchies introducing multi-association relationships, and supports imprecise measures.","PeriodicalId":175955,"journal":{"name":"2008 IEEE International Conference on Data Mining Workshops","volume":"93 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134158833","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}
S. Durbha, R. King, N. Younan, Santosh A. Rajender, Shruthi Bheemireddy
Coastal buoys and stations provide frequent, high quality marine observations for oceanographic study, weather service, atmospheric and public safety. Sharing of the generated data sets requires tremendous efforts and coordination among the different sensor network agencies to come to a shared understanding and for dissemination in a uniform way. Syntactic standardization provides data description models that are agreed upon by all the stakeholders. In addition, there is a need for semantic enrichment of the information sources which would help to understand the context of the data and helps to resolve the meaning, interpretation or usage of the same or related data. The standardized data models facilitate improved information retrieval on a variety of Spatiotemporal scales. In this paper we describe the mining of these information sources through a Web services based framework. The sensor observation service component of this framework allows operations such as spatial, temporal subsetting, filtering etc. Further, the terminology involved in the coastal domain is being conceptualized in the form of Ontology. The knowledgebase being developed using this ontological model is amenable to querying using SPARQL which is a standardized RDF query language. The knowledge-enabled client being developed will allow to process queries on the coastal sensors networks that goes beyond the prevalent key words based searches.
{"title":"Standards-Based Coastal Sensor Web","authors":"S. Durbha, R. King, N. Younan, Santosh A. Rajender, Shruthi Bheemireddy","doi":"10.1109/ICDMW.2008.131","DOIUrl":"https://doi.org/10.1109/ICDMW.2008.131","url":null,"abstract":"Coastal buoys and stations provide frequent, high quality marine observations for oceanographic study, weather service, atmospheric and public safety. Sharing of the generated data sets requires tremendous efforts and coordination among the different sensor network agencies to come to a shared understanding and for dissemination in a uniform way. Syntactic standardization provides data description models that are agreed upon by all the stakeholders. In addition, there is a need for semantic enrichment of the information sources which would help to understand the context of the data and helps to resolve the meaning, interpretation or usage of the same or related data. The standardized data models facilitate improved information retrieval on a variety of Spatiotemporal scales. In this paper we describe the mining of these information sources through a Web services based framework. The sensor observation service component of this framework allows operations such as spatial, temporal subsetting, filtering etc. Further, the terminology involved in the coastal domain is being conceptualized in the form of Ontology. The knowledgebase being developed using this ontological model is amenable to querying using SPARQL which is a standardized RDF query language. The knowledge-enabled client being developed will allow to process queries on the coastal sensors networks that goes beyond the prevalent key words based searches.","PeriodicalId":175955,"journal":{"name":"2008 IEEE International Conference on Data Mining Workshops","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130354133","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}
Depin Chen, Jun Yan, Gang Wang, Yan Xiong, Weiguo Fan, Zheng Chen
Recently, learning to rank technique has attracted much attention. However, the lack of labeled training data seriously limits its application in real-world tasks. In this paper, we propose to break this bottleneck by considering the cross-domain ldquotransfer of rank learningrdquo problem. Simultaneously, we propose a novel algorithm called TransRank, which can effectively utilize the labeled data from a source domain to enhance the learning of ranking function in the target domain. The proposed algorithm consists of three key steps. Firstly, we introduce a utility function to select the k-best queries from the source domain labeled data. Secondly, feature augmentation is performed on both source and target domain data, which can straightly adapt the ranking information from source domain to target domain. Finally, we utilize the classical ranking SVM to learn the enhanced ranking function on the augmented features. Experimental results on benchmark datasets well validate our proposed TransRank algorithm.
{"title":"TransRank: A Novel Algorithm for Transfer of Rank Learning","authors":"Depin Chen, Jun Yan, Gang Wang, Yan Xiong, Weiguo Fan, Zheng Chen","doi":"10.1109/ICDMW.2008.42","DOIUrl":"https://doi.org/10.1109/ICDMW.2008.42","url":null,"abstract":"Recently, learning to rank technique has attracted much attention. However, the lack of labeled training data seriously limits its application in real-world tasks. In this paper, we propose to break this bottleneck by considering the cross-domain ldquotransfer of rank learningrdquo problem. Simultaneously, we propose a novel algorithm called TransRank, which can effectively utilize the labeled data from a source domain to enhance the learning of ranking function in the target domain. The proposed algorithm consists of three key steps. Firstly, we introduce a utility function to select the k-best queries from the source domain labeled data. Secondly, feature augmentation is performed on both source and target domain data, which can straightly adapt the ranking information from source domain to target domain. Finally, we utilize the classical ranking SVM to learn the enhanced ranking function on the augmented features. Experimental results on benchmark datasets well validate our proposed TransRank algorithm.","PeriodicalId":175955,"journal":{"name":"2008 IEEE International Conference on Data Mining Workshops","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117320684","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}
As video summarization techniques have attracted more and more attention for efficient multimedia data management, quality assessment of video summary is required. To address the lack of automatic evaluation techniques, this paper proposes a novel framework including several new algorithms to assess the quality of the video summary against a given reference. First, we partition the reference video summary and the candidate video summary into the sequences of summary unit (SU). Then, we utilize alignment based algorithm to match the SUs in the candidate summary with the SUs in the corresponding reference summary. Third, we propose a novel similarity based 4 C - assessment algorithm to evaluate the candidate video summary from the perspective of coverage, conciseness, coherence, and context, respectively. Finally, the individual assessment results are integrated according to userpsilas requirement by a learning based weight adaptation method. The proposed framework and techniques are experimented on a standard dataset of TRECVID 2007 and show the good performance in automatic video summary assessment.
{"title":"Full-Reference Quality Assessment for Video Summary","authors":"Tongwei Ren, Yan Liu, Gangshan Wu","doi":"10.1109/ICDMW.2008.55","DOIUrl":"https://doi.org/10.1109/ICDMW.2008.55","url":null,"abstract":"As video summarization techniques have attracted more and more attention for efficient multimedia data management, quality assessment of video summary is required. To address the lack of automatic evaluation techniques, this paper proposes a novel framework including several new algorithms to assess the quality of the video summary against a given reference. First, we partition the reference video summary and the candidate video summary into the sequences of summary unit (SU). Then, we utilize alignment based algorithm to match the SUs in the candidate summary with the SUs in the corresponding reference summary. Third, we propose a novel similarity based 4 C - assessment algorithm to evaluate the candidate video summary from the perspective of coverage, conciseness, coherence, and context, respectively. Finally, the individual assessment results are integrated according to userpsilas requirement by a learning based weight adaptation method. The proposed framework and techniques are experimented on a standard dataset of TRECVID 2007 and show the good performance in automatic video summary assessment.","PeriodicalId":175955,"journal":{"name":"2008 IEEE International Conference on Data Mining Workshops","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122656943","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}
Automatic indexing of music by instruments and their types is a challenging problem, especially when multiple instruments are playing at the same time. We have built a database containing more than one million of music instrument sounds, each described by a large number o features including standard MPEG7 audio descriptors, features for speech recognition, and many new audio features developed by our team. Our previous research results show that all these features only lead to classifiers which successfully identify music instruments in monophonic music (only one instrument playing at a time). Their confidence for polyphonic music is much lower. This brought the need for blind sound source separation algorithms. In this paper, we present a new spectrum clustering enhanced method which improves the estimation of fundamental frequency as well as the balance of the categorization tree of training datasets, and therefore enhances the precision of automatic indexing. The system is recursively detecting the pitch of the predominant sound source, then calculates the features based on the estimated pitch, and then predicts the most similar spectrum by the corresponding classification tree, and finally subtracts the estimated predominant spectrum until silence is detected.
{"title":"Harmonic Blind Sound Source Isolation Enhanced by Spectrum Clustering","authors":"Cynthia Xin Zhang, Wenxin Jiang, Z. Ras","doi":"10.1109/ICDMW.2008.67","DOIUrl":"https://doi.org/10.1109/ICDMW.2008.67","url":null,"abstract":"Automatic indexing of music by instruments and their types is a challenging problem, especially when multiple instruments are playing at the same time. We have built a database containing more than one million of music instrument sounds, each described by a large number o features including standard MPEG7 audio descriptors, features for speech recognition, and many new audio features developed by our team. Our previous research results show that all these features only lead to classifiers which successfully identify music instruments in monophonic music (only one instrument playing at a time). Their confidence for polyphonic music is much lower. This brought the need for blind sound source separation algorithms. In this paper, we present a new spectrum clustering enhanced method which improves the estimation of fundamental frequency as well as the balance of the categorization tree of training datasets, and therefore enhances the precision of automatic indexing. The system is recursively detecting the pitch of the predominant sound source, then calculates the features based on the estimated pitch, and then predicts the most similar spectrum by the corresponding classification tree, and finally subtracts the estimated predominant spectrum until silence is detected.","PeriodicalId":175955,"journal":{"name":"2008 IEEE International Conference on Data Mining Workshops","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130378482","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}
Constraint-based mining has been proven to be extremely useful. It has been applied not only to many pattern discovery settings (e.g., for sequential pattern mining) but also, recently, on classification and clustering tasks (see, e.g., ). It appears as a key technology for an inductive database perspective on knowledge discovery in databases (KDD), and constraint-based mining is indeed an answer to important data mining issues (e.g., for supporting a priori relevancy and subjective interestingness but also to achieve computational feasibility). However, few authors study the nature of constraints and their semantics. Considering several examples of non trivial KDD processes, we discuss the Hows, Whys, and Whens of constraints in a broader context than. Our thesis is that most of the typical data mining methods are constraint-based techniques and that it is worth studying and designing them as such. In many cases, we exploit constraints that are not really explicit (e.g., the objective function optimization of a clustering for a given similarity measure) and/or constraints whose operational semantics are relaxed w.r.t. their declarative counterparts (e.g., the optimization constraint is not enforced because of some local optimization heuristics). We think that is important to explicit every primitive constraint and the operators that combine them because this constitutes the declarative semantics of the constraints and thus the mining queries. Then, a well-studied challenge is to design some operational semantics like correct and complete solvers and/or relaxation schemes for more or less complex constraints. Designing complete solvers has been extensively studied in useful but yet limited settings (see, e.g., the algorithms for exploiting combinations of monotonic and anti-monotonic primitives). It is however clear that many relevant constraints lack from such nice properties. On another hand, understanding constraint relaxation strategies remains fairly open, certainly because of its intrinsically heuristic nature. Interestingly, the recent approaches that suggest global pattern or model construction based on local patterns enable to revisit the relaxation issue thanks to constraint back propagation possibilities. This can be discussed within a case study on constrained co-clustering.
{"title":"If Constraint-Based Mining is the Answer: What is the Constraint? (Invited Talk)","authors":"Jean-François Boulicaut","doi":"10.1109/ICDMW.2008.96","DOIUrl":"https://doi.org/10.1109/ICDMW.2008.96","url":null,"abstract":"Constraint-based mining has been proven to be extremely useful. It has been applied not only to many pattern discovery settings (e.g., for sequential pattern mining) but also, recently, on classification and clustering tasks (see, e.g., ). It appears as a key technology for an inductive database perspective on knowledge discovery in databases (KDD), and constraint-based mining is indeed an answer to important data mining issues (e.g., for supporting a priori relevancy and subjective interestingness but also to achieve computational feasibility). However, few authors study the nature of constraints and their semantics. Considering several examples of non trivial KDD processes, we discuss the Hows, Whys, and Whens of constraints in a broader context than. Our thesis is that most of the typical data mining methods are constraint-based techniques and that it is worth studying and designing them as such. In many cases, we exploit constraints that are not really explicit (e.g., the objective function optimization of a clustering for a given similarity measure) and/or constraints whose operational semantics are relaxed w.r.t. their declarative counterparts (e.g., the optimization constraint is not enforced because of some local optimization heuristics). We think that is important to explicit every primitive constraint and the operators that combine them because this constitutes the declarative semantics of the constraints and thus the mining queries. Then, a well-studied challenge is to design some operational semantics like correct and complete solvers and/or relaxation schemes for more or less complex constraints. Designing complete solvers has been extensively studied in useful but yet limited settings (see, e.g., the algorithms for exploiting combinations of monotonic and anti-monotonic primitives). It is however clear that many relevant constraints lack from such nice properties. On another hand, understanding constraint relaxation strategies remains fairly open, certainly because of its intrinsically heuristic nature. Interestingly, the recent approaches that suggest global pattern or model construction based on local patterns enable to revisit the relaxation issue thanks to constraint back propagation possibilities. This can be discussed within a case study on constrained co-clustering.","PeriodicalId":175955,"journal":{"name":"2008 IEEE International Conference on Data Mining Workshops","volume":"196 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131183480","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}
Christoph Weyerhaeuser, Tobias Mindnich, Franz Färber, Wolfgang Lehner
Within business Intelligence contexts, the importance of data mining algorithms is continuously increasing, particularly from the perspective of applications and users that demand novel algorithms on the one hand and an efficient implementation exploiting novel system architectures on the other hand. Within this paper, we focus on the latter issue and report our experience with the exploitation of graphic card processor technology within the SAP NetWeaver business intelligence accelerator (BIA). The BIA represents a highly distributed analytical engine that supports OLAP and data mining processing primitives. The system organizes data entities in column-wise fashion and its operation is completely main-memory-based. Since case studies have shown that classic data mining queries spend a large portion of their runtime on scanning and filtering the data as a necessary prerequisite to the actual mining step, our main goal was to speed up this expensive scanning and filtering process. In a first step, the paper outlines the basic data mining processing techniques within SAP NetWeaver BIA and illustrates the implementation of scans and filters. In a second step, we give insight into the main features of a hybrid system architecture design exploiting graphic card processor technology. Finally, we sketch the implementation and give details of our vast evaluations.
{"title":"Exploiting Graphic Card Processor Technology to Accelerate Data Mining Queries in SAP NetWeaver BIA","authors":"Christoph Weyerhaeuser, Tobias Mindnich, Franz Färber, Wolfgang Lehner","doi":"10.1109/ICDMW.2008.61","DOIUrl":"https://doi.org/10.1109/ICDMW.2008.61","url":null,"abstract":"Within business Intelligence contexts, the importance of data mining algorithms is continuously increasing, particularly from the perspective of applications and users that demand novel algorithms on the one hand and an efficient implementation exploiting novel system architectures on the other hand. Within this paper, we focus on the latter issue and report our experience with the exploitation of graphic card processor technology within the SAP NetWeaver business intelligence accelerator (BIA). The BIA represents a highly distributed analytical engine that supports OLAP and data mining processing primitives. The system organizes data entities in column-wise fashion and its operation is completely main-memory-based. Since case studies have shown that classic data mining queries spend a large portion of their runtime on scanning and filtering the data as a necessary prerequisite to the actual mining step, our main goal was to speed up this expensive scanning and filtering process. In a first step, the paper outlines the basic data mining processing techniques within SAP NetWeaver BIA and illustrates the implementation of scans and filters. In a second step, we give insight into the main features of a hybrid system architecture design exploiting graphic card processor technology. Finally, we sketch the implementation and give details of our vast evaluations.","PeriodicalId":175955,"journal":{"name":"2008 IEEE International Conference on Data Mining Workshops","volume":"15 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125114658","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}
Research on streaming data classification has been mostly based on the assumption that data can be fully labelled. However, this is impractical. Firstly it is impossible to make a complete labelling before all data has arrived. Secondly it is generally very expensive to obtain fully labelled data by using man power. Thirdly user interests may change with time so the labels issued earlier may be inconsistent with the labels issued later - this represents concept drift. In this paper, we consider the problem of one-class classification on text stream with respect to concept drift where a large volume of documents arrives at a high speed and with change of user interests and data distribution. In this case, only a small number of positively labelled documents is available for training. We propose a stacking style ensemble-based approach and have compared it to all other window-based approaches, such as single window, fixed window, and full memory approaches. Our experiment results demonstrate that the proposed ensemble approach outperforms all other approaches.
{"title":"One-Class Classification of Text Streams with Concept Drift","authors":"Yang Zhang, Xue Li, M. Orlowska","doi":"10.1109/ICDMW.2008.54","DOIUrl":"https://doi.org/10.1109/ICDMW.2008.54","url":null,"abstract":"Research on streaming data classification has been mostly based on the assumption that data can be fully labelled. However, this is impractical. Firstly it is impossible to make a complete labelling before all data has arrived. Secondly it is generally very expensive to obtain fully labelled data by using man power. Thirdly user interests may change with time so the labels issued earlier may be inconsistent with the labels issued later - this represents concept drift. In this paper, we consider the problem of one-class classification on text stream with respect to concept drift where a large volume of documents arrives at a high speed and with change of user interests and data distribution. In this case, only a small number of positively labelled documents is available for training. We propose a stacking style ensemble-based approach and have compared it to all other window-based approaches, such as single window, fixed window, and full memory approaches. Our experiment results demonstrate that the proposed ensemble approach outperforms all other approaches.","PeriodicalId":175955,"journal":{"name":"2008 IEEE International Conference on Data Mining Workshops","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126869782","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}
A new feature description is used for human behaviour representation and recognition. The feature is based on Radon transforms of extracted silhouettes. Key postures are selected based on the Radon transform. Key postures are combined to construct an action template for each sequence. Linear discriminant analysis (LDA) is applied to the set of key postures to obtain low dimensional feature vectors. Different classification methods are used to classify each sequence. Experiments are carried out based on a publically available human behaviour database and the results are exciting.
{"title":"Human Action Recognition by Radon Transform","authors":"Yan Chen, Qiang Wu, Xiangjian He","doi":"10.1109/ICDMW.2008.26","DOIUrl":"https://doi.org/10.1109/ICDMW.2008.26","url":null,"abstract":"A new feature description is used for human behaviour representation and recognition. The feature is based on Radon transforms of extracted silhouettes. Key postures are selected based on the Radon transform. Key postures are combined to construct an action template for each sequence. Linear discriminant analysis (LDA) is applied to the set of key postures to obtain low dimensional feature vectors. Different classification methods are used to classify each sequence. Experiments are carried out based on a publically available human behaviour database and the results are exciting.","PeriodicalId":175955,"journal":{"name":"2008 IEEE International Conference on Data Mining Workshops","volume":"58 19","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113933429","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 decision tree-based classification is a popular approach for pattern recognition and data mining. Most decision tree induction methods assume training data being present at one central location. Given the growth in distributed databases at geographically dispersed locations, the methods for decision tree induction in distributed settings are gaining importance. This paper describes one distributed learning algorithm which extends the original(centralized) CHAID algorithm to its distributed version. This distributed algorithm generates exactly the same results as its centralized counterpart. For completeness, a distributed quantization method is proposed so that continuous data can be processed by our algorithm. Experimental results for several well known data sets are presented and compared with decision trees generated using CHAID with centrally stored data.
{"title":"Chi-Square Test Based Decision Trees Induction in Distributed Environment","authors":"Jie Ouyang, Nilesh V. Patel, I. Sethi","doi":"10.1109/ICDMW.2008.37","DOIUrl":"https://doi.org/10.1109/ICDMW.2008.37","url":null,"abstract":"The decision tree-based classification is a popular approach for pattern recognition and data mining. Most decision tree induction methods assume training data being present at one central location. Given the growth in distributed databases at geographically dispersed locations, the methods for decision tree induction in distributed settings are gaining importance. This paper describes one distributed learning algorithm which extends the original(centralized) CHAID algorithm to its distributed version. This distributed algorithm generates exactly the same results as its centralized counterpart. For completeness, a distributed quantization method is proposed so that continuous data can be processed by our algorithm. Experimental results for several well known data sets are presented and compared with decision trees generated using CHAID with centrally stored data.","PeriodicalId":175955,"journal":{"name":"2008 IEEE International Conference on Data Mining Workshops","volume":"100 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114501374","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：