The Experts below are selected from a list of 132 Experts worldwide ranked by ideXlab platform
Koichi Furukawa - One of the best experts on this subject based on the ideXlab platform.
-
discovering exceptional information from Customer Inquiry by association rule miner
Discovery Science, 2003Co-Authors: Keiko Shimazu, Atsuhito Momma, Koichi FurukawaAbstract:This paper reports the results of our experimental study on a new method of applying an association rule miner to discover useful information from a text database. It has been claimed that association rule mining is not suited for text mining. To overcome this problem, we propose (1) to generate a sequential data set of words with dependency structure from a Japanese text database, and (2) to employ a new method for extracting meaningful association rules by applying a new rule selection criterion. Each Inquiry was converted to a list of word pairs, having dependency relationship in the original sentence. The association rules were acquired regarding each pair of words as an item. The rule selection criterion derived from our principle of giving heavier weights to co-occurrence of multiple items than to single item occurrence. We regarded a rule as important if the existence of the items in the rule body significantly affected the occurrence of the item in the rule head. Based on this method, we conducted experiments on a Customer Inquiry database in a call center of a company and successfully acquired practical meaningful rules, which were not too general nor appeared only rarely. Also, they were not acquired by only simple keyword retrieval. Additionally, inquiries with multiple aspects were properly classified into corresponding multiple categories. Furthermore, we compared (i) rules obtained from a sequential data set of words with dependency structure, which we propose in this paper, and those without dependency structure, as well as (ii) rules acquired through the association rule selection criterion and those through the conventional criteria. As a result, discovery of meaningful rules increased 14.3-fold in the first comparison, and we confirmed that our criterion enables to obtain rules according to the objectives more precisely in the second comparison.
-
experimental study of discovering essential information from Customer Inquiry
Knowledge Discovery and Data Mining, 2003Co-Authors: Keiko Shimazu, Atsuhito Momma, Koichi FurukawaAbstract:This paper reports the result of our experimental study on a new method of applying an association rule miner to discover useful information from Customer Inquiry database in a call center of a company. It has been claimed that association rule mining is not suited for text mining. To overcome this problem, we propose (1) to generate sequential data set of words with dependency structure from the Japanese text database, and (2) to employ a new method for extracting meaningful association rules by applying a new rule selection criterion. Each Inquiry in the sequential data was represented as a list of word pairs, each of which consists of a verb and its dependent noun. The association rules were induced regarding each pair of words as an item. The rule selection criterion comes from our principle that we put heavier weights to co-occurrence of multiple items more than single item occurrence. We regarded a rule important if the existence of the items in the rule body significantly affects the occurrence of the item in the rule head. The selected rules were then categorized to form meaningful information classes. With this method, we succeeded in extracting useful information classes from the text database, which were not acquired by only simple keyword retrieval. Also, inquiries with multiple aspects were properly classified into corresponding multiple categories.
-
KDD - Experimental study of discovering essential information from Customer Inquiry
Proceedings of the ninth ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '03, 2003Co-Authors: Keiko Shimazu, Atsuhito Momma, Koichi FurukawaAbstract:This paper reports the result of our experimental study on a new method of applying an association rule miner to discover useful information from Customer Inquiry database in a call center of a company. It has been claimed that association rule mining is not suited for text mining. To overcome this problem, we propose (1) to generate sequential data set of words with dependency structure from the Japanese text database, and (2) to employ a new method for extracting meaningful association rules by applying a new rule selection criterion. Each Inquiry in the sequential data was represented as a list of word pairs, each of which consists of a verb and its dependent noun. The association rules were induced regarding each pair of words as an item. The rule selection criterion comes from our principle that we put heavier weights to co-occurrence of multiple items more than single item occurrence. We regarded a rule important if the existence of the items in the rule body significantly affects the occurrence of the item in the rule head. The selected rules were then categorized to form meaningful information classes. With this method, we succeeded in extracting useful information classes from the text database, which were not acquired by only simple keyword retrieval. Also, inquiries with multiple aspects were properly classified into corresponding multiple categories.
-
Discovery Science - Discovering Exceptional Information from Customer Inquiry by Association Rule Miner
Discovery Science, 2003Co-Authors: Keiko Shimazu, Atsuhito Momma, Koichi FurukawaAbstract:This paper reports the results of our experimental study on a new method of applying an association rule miner to discover useful information from a text database. It has been claimed that association rule mining is not suited for text mining. To overcome this problem, we propose (1) to generate a sequential data set of words with dependency structure from a Japanese text database, and (2) to employ a new method for extracting meaningful association rules by applying a new rule selection criterion. Each Inquiry was converted to a list of word pairs, having dependency relationship in the original sentence. The association rules were acquired regarding each pair of words as an item. The rule selection criterion derived from our principle of giving heavier weights to co-occurrence of multiple items than to single item occurrence. We regarded a rule as important if the existence of the items in the rule body significantly affected the occurrence of the item in the rule head. Based on this method, we conducted experiments on a Customer Inquiry database in a call center of a company and successfully acquired practical meaningful rules, which were not too general nor appeared only rarely. Also, they were not acquired by only simple keyword retrieval. Additionally, inquiries with multiple aspects were properly classified into corresponding multiple categories. Furthermore, we compared (i) rules obtained from a sequential data set of words with dependency structure, which we propose in this paper, and those without dependency structure, as well as (ii) rules acquired through the association rule selection criterion and those through the conventional criteria. As a result, discovery of meaningful rules increased 14.3-fold in the first comparison, and we confirmed that our criterion enables to obtain rules according to the objectives more precisely in the second comparison.
Keiko Shimazu - One of the best experts on this subject based on the ideXlab platform.
-
discovering exceptional information from Customer Inquiry by association rule miner
Discovery Science, 2003Co-Authors: Keiko Shimazu, Atsuhito Momma, Koichi FurukawaAbstract:This paper reports the results of our experimental study on a new method of applying an association rule miner to discover useful information from a text database. It has been claimed that association rule mining is not suited for text mining. To overcome this problem, we propose (1) to generate a sequential data set of words with dependency structure from a Japanese text database, and (2) to employ a new method for extracting meaningful association rules by applying a new rule selection criterion. Each Inquiry was converted to a list of word pairs, having dependency relationship in the original sentence. The association rules were acquired regarding each pair of words as an item. The rule selection criterion derived from our principle of giving heavier weights to co-occurrence of multiple items than to single item occurrence. We regarded a rule as important if the existence of the items in the rule body significantly affected the occurrence of the item in the rule head. Based on this method, we conducted experiments on a Customer Inquiry database in a call center of a company and successfully acquired practical meaningful rules, which were not too general nor appeared only rarely. Also, they were not acquired by only simple keyword retrieval. Additionally, inquiries with multiple aspects were properly classified into corresponding multiple categories. Furthermore, we compared (i) rules obtained from a sequential data set of words with dependency structure, which we propose in this paper, and those without dependency structure, as well as (ii) rules acquired through the association rule selection criterion and those through the conventional criteria. As a result, discovery of meaningful rules increased 14.3-fold in the first comparison, and we confirmed that our criterion enables to obtain rules according to the objectives more precisely in the second comparison.
-
experimental study of discovering essential information from Customer Inquiry
Knowledge Discovery and Data Mining, 2003Co-Authors: Keiko Shimazu, Atsuhito Momma, Koichi FurukawaAbstract:This paper reports the result of our experimental study on a new method of applying an association rule miner to discover useful information from Customer Inquiry database in a call center of a company. It has been claimed that association rule mining is not suited for text mining. To overcome this problem, we propose (1) to generate sequential data set of words with dependency structure from the Japanese text database, and (2) to employ a new method for extracting meaningful association rules by applying a new rule selection criterion. Each Inquiry in the sequential data was represented as a list of word pairs, each of which consists of a verb and its dependent noun. The association rules were induced regarding each pair of words as an item. The rule selection criterion comes from our principle that we put heavier weights to co-occurrence of multiple items more than single item occurrence. We regarded a rule important if the existence of the items in the rule body significantly affects the occurrence of the item in the rule head. The selected rules were then categorized to form meaningful information classes. With this method, we succeeded in extracting useful information classes from the text database, which were not acquired by only simple keyword retrieval. Also, inquiries with multiple aspects were properly classified into corresponding multiple categories.
-
KDD - Experimental study of discovering essential information from Customer Inquiry
Proceedings of the ninth ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '03, 2003Co-Authors: Keiko Shimazu, Atsuhito Momma, Koichi FurukawaAbstract:This paper reports the result of our experimental study on a new method of applying an association rule miner to discover useful information from Customer Inquiry database in a call center of a company. It has been claimed that association rule mining is not suited for text mining. To overcome this problem, we propose (1) to generate sequential data set of words with dependency structure from the Japanese text database, and (2) to employ a new method for extracting meaningful association rules by applying a new rule selection criterion. Each Inquiry in the sequential data was represented as a list of word pairs, each of which consists of a verb and its dependent noun. The association rules were induced regarding each pair of words as an item. The rule selection criterion comes from our principle that we put heavier weights to co-occurrence of multiple items more than single item occurrence. We regarded a rule important if the existence of the items in the rule body significantly affects the occurrence of the item in the rule head. The selected rules were then categorized to form meaningful information classes. With this method, we succeeded in extracting useful information classes from the text database, which were not acquired by only simple keyword retrieval. Also, inquiries with multiple aspects were properly classified into corresponding multiple categories.
-
Discovery Science - Discovering Exceptional Information from Customer Inquiry by Association Rule Miner
Discovery Science, 2003Co-Authors: Keiko Shimazu, Atsuhito Momma, Koichi FurukawaAbstract:This paper reports the results of our experimental study on a new method of applying an association rule miner to discover useful information from a text database. It has been claimed that association rule mining is not suited for text mining. To overcome this problem, we propose (1) to generate a sequential data set of words with dependency structure from a Japanese text database, and (2) to employ a new method for extracting meaningful association rules by applying a new rule selection criterion. Each Inquiry was converted to a list of word pairs, having dependency relationship in the original sentence. The association rules were acquired regarding each pair of words as an item. The rule selection criterion derived from our principle of giving heavier weights to co-occurrence of multiple items than to single item occurrence. We regarded a rule as important if the existence of the items in the rule body significantly affected the occurrence of the item in the rule head. Based on this method, we conducted experiments on a Customer Inquiry database in a call center of a company and successfully acquired practical meaningful rules, which were not too general nor appeared only rarely. Also, they were not acquired by only simple keyword retrieval. Additionally, inquiries with multiple aspects were properly classified into corresponding multiple categories. Furthermore, we compared (i) rules obtained from a sequential data set of words with dependency structure, which we propose in this paper, and those without dependency structure, as well as (ii) rules acquired through the association rule selection criterion and those through the conventional criteria. As a result, discovery of meaningful rules increased 14.3-fold in the first comparison, and we confirmed that our criterion enables to obtain rules according to the objectives more precisely in the second comparison.
Atsuhito Momma - One of the best experts on this subject based on the ideXlab platform.
-
discovering exceptional information from Customer Inquiry by association rule miner
Discovery Science, 2003Co-Authors: Keiko Shimazu, Atsuhito Momma, Koichi FurukawaAbstract:This paper reports the results of our experimental study on a new method of applying an association rule miner to discover useful information from a text database. It has been claimed that association rule mining is not suited for text mining. To overcome this problem, we propose (1) to generate a sequential data set of words with dependency structure from a Japanese text database, and (2) to employ a new method for extracting meaningful association rules by applying a new rule selection criterion. Each Inquiry was converted to a list of word pairs, having dependency relationship in the original sentence. The association rules were acquired regarding each pair of words as an item. The rule selection criterion derived from our principle of giving heavier weights to co-occurrence of multiple items than to single item occurrence. We regarded a rule as important if the existence of the items in the rule body significantly affected the occurrence of the item in the rule head. Based on this method, we conducted experiments on a Customer Inquiry database in a call center of a company and successfully acquired practical meaningful rules, which were not too general nor appeared only rarely. Also, they were not acquired by only simple keyword retrieval. Additionally, inquiries with multiple aspects were properly classified into corresponding multiple categories. Furthermore, we compared (i) rules obtained from a sequential data set of words with dependency structure, which we propose in this paper, and those without dependency structure, as well as (ii) rules acquired through the association rule selection criterion and those through the conventional criteria. As a result, discovery of meaningful rules increased 14.3-fold in the first comparison, and we confirmed that our criterion enables to obtain rules according to the objectives more precisely in the second comparison.
-
experimental study of discovering essential information from Customer Inquiry
Knowledge Discovery and Data Mining, 2003Co-Authors: Keiko Shimazu, Atsuhito Momma, Koichi FurukawaAbstract:This paper reports the result of our experimental study on a new method of applying an association rule miner to discover useful information from Customer Inquiry database in a call center of a company. It has been claimed that association rule mining is not suited for text mining. To overcome this problem, we propose (1) to generate sequential data set of words with dependency structure from the Japanese text database, and (2) to employ a new method for extracting meaningful association rules by applying a new rule selection criterion. Each Inquiry in the sequential data was represented as a list of word pairs, each of which consists of a verb and its dependent noun. The association rules were induced regarding each pair of words as an item. The rule selection criterion comes from our principle that we put heavier weights to co-occurrence of multiple items more than single item occurrence. We regarded a rule important if the existence of the items in the rule body significantly affects the occurrence of the item in the rule head. The selected rules were then categorized to form meaningful information classes. With this method, we succeeded in extracting useful information classes from the text database, which were not acquired by only simple keyword retrieval. Also, inquiries with multiple aspects were properly classified into corresponding multiple categories.
-
KDD - Experimental study of discovering essential information from Customer Inquiry
Proceedings of the ninth ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '03, 2003Co-Authors: Keiko Shimazu, Atsuhito Momma, Koichi FurukawaAbstract:This paper reports the result of our experimental study on a new method of applying an association rule miner to discover useful information from Customer Inquiry database in a call center of a company. It has been claimed that association rule mining is not suited for text mining. To overcome this problem, we propose (1) to generate sequential data set of words with dependency structure from the Japanese text database, and (2) to employ a new method for extracting meaningful association rules by applying a new rule selection criterion. Each Inquiry in the sequential data was represented as a list of word pairs, each of which consists of a verb and its dependent noun. The association rules were induced regarding each pair of words as an item. The rule selection criterion comes from our principle that we put heavier weights to co-occurrence of multiple items more than single item occurrence. We regarded a rule important if the existence of the items in the rule body significantly affects the occurrence of the item in the rule head. The selected rules were then categorized to form meaningful information classes. With this method, we succeeded in extracting useful information classes from the text database, which were not acquired by only simple keyword retrieval. Also, inquiries with multiple aspects were properly classified into corresponding multiple categories.
-
Discovery Science - Discovering Exceptional Information from Customer Inquiry by Association Rule Miner
Discovery Science, 2003Co-Authors: Keiko Shimazu, Atsuhito Momma, Koichi FurukawaAbstract:This paper reports the results of our experimental study on a new method of applying an association rule miner to discover useful information from a text database. It has been claimed that association rule mining is not suited for text mining. To overcome this problem, we propose (1) to generate a sequential data set of words with dependency structure from a Japanese text database, and (2) to employ a new method for extracting meaningful association rules by applying a new rule selection criterion. Each Inquiry was converted to a list of word pairs, having dependency relationship in the original sentence. The association rules were acquired regarding each pair of words as an item. The rule selection criterion derived from our principle of giving heavier weights to co-occurrence of multiple items than to single item occurrence. We regarded a rule as important if the existence of the items in the rule body significantly affected the occurrence of the item in the rule head. Based on this method, we conducted experiments on a Customer Inquiry database in a call center of a company and successfully acquired practical meaningful rules, which were not too general nor appeared only rarely. Also, they were not acquired by only simple keyword retrieval. Additionally, inquiries with multiple aspects were properly classified into corresponding multiple categories. Furthermore, we compared (i) rules obtained from a sequential data set of words with dependency structure, which we propose in this paper, and those without dependency structure, as well as (ii) rules acquired through the association rule selection criterion and those through the conventional criteria. As a result, discovery of meaningful rules increased 14.3-fold in the first comparison, and we confirmed that our criterion enables to obtain rules according to the objectives more precisely in the second comparison.
Zekun Yang - One of the best experts on this subject based on the ideXlab platform.
-
non negative matrix factorization for overlapping clustering of Customer Inquiry and review data
International Conference on Machine Learning, 2018Co-Authors: Zekun YangAbstract:Considering the complexity of clustering text datasets in terms of informal user generated content and the fact that there are multiple labels for each data point in many informal user generated content datasets, this paper focuses on Non-negative Matrix Factorization (NMF) algorithms for Overlapping Clustering of Customer Inquiry and review data, which has seldom been discussed in previous literature. We extend the use of Semi-NMF and Convex-NMF to Overlapping Clustering and develop a procedure of applying SemiNMF and Convex-NMF on Overlapping Clustering of text data. The developed procedure is tested based on Customer review and Inquiry datasets. The results of comparing SemiNMF and Convex-NMF with a baseline model demonstrate that they have advantages over the baseline model, since they do not need to adjust parameters to obtain similarly strong clustering performances. Moreover, we compare different methods of picking labels for generating Overlapping Clustering results from Soft Clustering algorithms, and it is concluded that thresholding by mean method is a simpler and relatively more reliable method compared to maximum n method.
-
ICMLC - Non-negative Matrix Factorization for Overlapping Clustering of Customer Inquiry and Review Data
Proceedings of the 2018 10th International Conference on Machine Learning and Computing, 2018Co-Authors: Zekun YangAbstract:Considering the complexity of clustering text datasets in terms of informal user generated content and the fact that there are multiple labels for each data point in many informal user generated content datasets, this paper focuses on Non-negative Matrix Factorization (NMF) algorithms for Overlapping Clustering of Customer Inquiry and review data, which has seldom been discussed in previous literature. We extend the use of Semi-NMF and Convex-NMF to Overlapping Clustering and develop a procedure of applying SemiNMF and Convex-NMF on Overlapping Clustering of text data. The developed procedure is tested based on Customer review and Inquiry datasets. The results of comparing SemiNMF and Convex-NMF with a baseline model demonstrate that they have advantages over the baseline model, since they do not need to adjust parameters to obtain similarly strong clustering performances. Moreover, we compare different methods of picking labels for generating Overlapping Clustering results from Soft Clustering algorithms, and it is concluded that thresholding by mean method is a simpler and relatively more reliable method compared to maximum n method.
Heinz Raufer - One of the best experts on this subject based on the ideXlab platform.
-
integrated document and workflow management applied to the offer processing of a machine tool company
Conference on Organizational Computing Systems, 1995Co-Authors: Stefan Morschheuser, Heinz RauferAbstract:Introducing document and workflow management systems causes two main problems: How can the supported business processes be adequately modeled? How can existing information systems and databases be integrated? Within this paper, we present tools, methods and other approaches, which are designed to solve these problems. They are illustrated by the offer processing of a machine tool company. This process starts with a Customer Inquiry for a particular product and finishes with a customized offer. Several levels of integrated document and workflow management are discussed in section 2. This is followed by an introduction to the current offer processing used by our partner company INA Waelzlager Schaeffler KG, in which we highlight its main weaknesses. The fourth section describes a newly developed document-oriented tool to model business processes, and which serves also as a means of analyzing the current offer processing. The subject matter of the next chapter is the prototypical realization of a document and workflow management system at INA, which also comprises integrated application programs. The objective consists of a “lean integration” in order to avoid methods like total IS-reengineering or the use of highly integrated, but rigid standard software, where these are unreasonable heavy for small to medium enterprises.
-
COOCS - Integrated document and workflow management applied to the offer processing of a machine tool company
Proceedings of conference on Organizational computing systems - COCS '95, 1995Co-Authors: Stefan Morschheuser, Heinz RauferAbstract:Introducing document and workflow management systems causes two main problems: How can the supported business processes be adequately modeled? How can existing information systems and databases be integrated? Within this paper, we present tools, methods and other approaches, which are designed to solve these problems. They are illustrated by the offer processing of a machine tool company. This process starts with a Customer Inquiry for a particular product and finishes with a customized offer. Several levels of integrated document and workflow management are discussed in section 2. This is followed by an introduction to the current offer processing used by our partner company INA Waelzlager Schaeffler KG, in which we highlight its main weaknesses. The fourth section describes a newly developed document-oriented tool to model business processes, and which serves also as a means of analyzing the current offer processing. The subject matter of the next chapter is the prototypical realization of a document and workflow management system at INA, which also comprises integrated application programs. The objective consists of a “lean integration” in order to avoid methods like total IS-reengineering or the use of highly integrated, but rigid standard software, where these are unreasonable heavy for small to medium enterprises.
