Peng Liu, Lei Lei, Junjie Yin, Wei Zhang, Wu Naijun, E. El-Darzi
Data mining approaches have been widely applied in the field of healthcare. At the same time it is recognized that most healthcare datasets are full of missing values. In this paper we apply decision trees, Naive Bayesian classifiers and feature selection methods to a geriatric hospital dataset in order to predict inpatient length of stay, especially for the long stay patients
{"title":"Healthcare Data Mining: Prediction Inpatient Length of Stay","authors":"Peng Liu, Lei Lei, Junjie Yin, Wei Zhang, Wu Naijun, E. El-Darzi","doi":"10.1109/IS.2006.348528","DOIUrl":"https://doi.org/10.1109/IS.2006.348528","url":null,"abstract":"Data mining approaches have been widely applied in the field of healthcare. At the same time it is recognized that most healthcare datasets are full of missing values. In this paper we apply decision trees, Naive Bayesian classifiers and feature selection methods to a geriatric hospital dataset in order to predict inpatient length of stay, especially for the long stay patients","PeriodicalId":116809,"journal":{"name":"2006 3rd International IEEE Conference Intelligent Systems","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124952234","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 the application of genetic algorithms (GAs) to the design of an intelligent switching surface for variable structure adaptive model following controller for higher order systems with unmodelled dynamics/parameter variations. The conventional approach for the design of switching surface by pole placement method often lead to large value of control signals. A method for obtaining an intelligent switching surface in a computationally efficient manner is proposed in this paper. The proposed method make use of GAs to evolve a switching surface which ensures minimum disruption of the poles when variations/uncertainties act on the system. If minimum disruption of the poles is not ensured, higher control signal will be required to maintain sliding mode motion. The proposed methodology is applied to a practical system namely a flexible one-link manipulator and the results obtained are compared to the results obtained by applying the conventional design. The comparison reveals the efficacy of the proposed method
{"title":"Intelligent Switching Surface for Variable Structure Adaptive Model Following Control","authors":"S. Thomas, H. Reddy","doi":"10.1109/IS.2006.348437","DOIUrl":"https://doi.org/10.1109/IS.2006.348437","url":null,"abstract":"This paper presents the application of genetic algorithms (GAs) to the design of an intelligent switching surface for variable structure adaptive model following controller for higher order systems with unmodelled dynamics/parameter variations. The conventional approach for the design of switching surface by pole placement method often lead to large value of control signals. A method for obtaining an intelligent switching surface in a computationally efficient manner is proposed in this paper. The proposed method make use of GAs to evolve a switching surface which ensures minimum disruption of the poles when variations/uncertainties act on the system. If minimum disruption of the poles is not ensured, higher control signal will be required to maintain sliding mode motion. The proposed methodology is applied to a practical system namely a flexible one-link manipulator and the results obtained are compared to the results obtained by applying the conventional design. The comparison reveals the efficacy of the proposed method","PeriodicalId":116809,"journal":{"name":"2006 3rd International IEEE Conference Intelligent Systems","volume":"82 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122161146","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}
Prostate cancer remains one of the leading causes of cancer death worldwide, with a reported incidence rate of 650,000 cases per annum worldwide. The causal factors of prostate cancer still remain to be determined. In this paper, we investigate a medical dataset containing clinical information on 502 prostate cancer patients using the machine learning technique of rough sets. Our preliminary results yield a classification accuracy of 90%, with high sensitivity and specificity (both at approximately 91%). Our results yield a predictive positive value (PPN) of 81% and a predictive negative value (PNV) of 95%. In addition to the high classification accuracy of our system, the rough set approach also provides a rule-based inference mechanism for information extraction that is suitable for integration into a rule-based system. The generated rules relate directly to the attributes and their values and provide a direct mapping between them
{"title":"Data Mining a Prostate Cancer Dataset Using Rough Sets","authors":"K. Revett, S.T. de Magalhaes, H. Santos","doi":"10.1109/IS.2006.348433","DOIUrl":"https://doi.org/10.1109/IS.2006.348433","url":null,"abstract":"Prostate cancer remains one of the leading causes of cancer death worldwide, with a reported incidence rate of 650,000 cases per annum worldwide. The causal factors of prostate cancer still remain to be determined. In this paper, we investigate a medical dataset containing clinical information on 502 prostate cancer patients using the machine learning technique of rough sets. Our preliminary results yield a classification accuracy of 90%, with high sensitivity and specificity (both at approximately 91%). Our results yield a predictive positive value (PPN) of 81% and a predictive negative value (PNV) of 95%. In addition to the high classification accuracy of our system, the rough set approach also provides a rule-based inference mechanism for information extraction that is suitable for integration into a rule-based system. The generated rules relate directly to the attributes and their values and provide a direct mapping between them","PeriodicalId":116809,"journal":{"name":"2006 3rd International IEEE Conference Intelligent Systems","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126077852","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 deals with a state observation problem when the dynamic model of a plant contains an uncertainty or it is completely unknown (the only some smoothness properties are assumed to be in force). The dynamic neural network approach is applied in this informative situation. A new learning law, containing relay (signum) terms, is suggested to be in use. The nominal parameters of this procedure are adjusted during the preliminary "training process" where the sliding-mode technique as well as the LS-method are applied to obtain the "best" nominal parameter values using training experimental data. The upper bounds for the weights as well as for the averaged estimation error are derived. Two numeric examples illustrate this approach: first, the water ozone-purification process supplied by a bilinear model with unknown parameters, and, second, a nonlinear mechanical system, governed by the Euler's equations with unknown parameters and noises
{"title":"Dynamic Neural Observer with Sliding Mode Learning","authors":"I. Chairez, A. Poznyak, T. Poznyak","doi":"10.1109/IS.2006.348487","DOIUrl":"https://doi.org/10.1109/IS.2006.348487","url":null,"abstract":"This paper deals with a state observation problem when the dynamic model of a plant contains an uncertainty or it is completely unknown (the only some smoothness properties are assumed to be in force). The dynamic neural network approach is applied in this informative situation. A new learning law, containing relay (signum) terms, is suggested to be in use. The nominal parameters of this procedure are adjusted during the preliminary \"training process\" where the sliding-mode technique as well as the LS-method are applied to obtain the \"best\" nominal parameter values using training experimental data. The upper bounds for the weights as well as for the averaged estimation error are derived. Two numeric examples illustrate this approach: first, the water ozone-purification process supplied by a bilinear model with unknown parameters, and, second, a nonlinear mechanical system, governed by the Euler's equations with unknown parameters and noises","PeriodicalId":116809,"journal":{"name":"2006 3rd International IEEE Conference Intelligent Systems","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127409430","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 decision support system based on rough sets and a probabilistic neural network is presented. Rough sets were employed as they have the capacity to reduce the dimensionality of the dataset and also produce a set of readily understandable rules. A probabilistic neural network was also employed to classify this dataset, comparing the classification accuracy to that obtained with rough sets. We firstly evaluate the effectiveness of these machine learning algorithms on a real-life small biomedical dataset. The classification results indicate that both classifiers produce a high level of accuracy (87% or better). The rough sets algorithm produced a set of rules that are readily interpretable by a domain expert. The PNN algorithm produced a classifier that was robust to noise and missing values. These preliminary results indicate that the both rough sets and PNN machine learning approaches can be successfully applied synergistically to biomedical datasets that contain a variety of attribute types, missing values and multiple decision classes
{"title":"Mining A Primary Biliary Cirrhosis Dataset Using Rough Sets and a Probabilistic Neural Network","authors":"K. Revett, F. Gorunescu, M. Gorunescu, M. Ene","doi":"10.1109/IS.2006.348432","DOIUrl":"https://doi.org/10.1109/IS.2006.348432","url":null,"abstract":"In this paper, a decision support system based on rough sets and a probabilistic neural network is presented. Rough sets were employed as they have the capacity to reduce the dimensionality of the dataset and also produce a set of readily understandable rules. A probabilistic neural network was also employed to classify this dataset, comparing the classification accuracy to that obtained with rough sets. We firstly evaluate the effectiveness of these machine learning algorithms on a real-life small biomedical dataset. The classification results indicate that both classifiers produce a high level of accuracy (87% or better). The rough sets algorithm produced a set of rules that are readily interpretable by a domain expert. The PNN algorithm produced a classifier that was robust to noise and missing values. These preliminary results indicate that the both rough sets and PNN machine learning approaches can be successfully applied synergistically to biomedical datasets that contain a variety of attribute types, missing values and multiple decision classes","PeriodicalId":116809,"journal":{"name":"2006 3rd International IEEE Conference Intelligent Systems","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121250074","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 discusses an evaluation method of fuzzy numbers/fuzzy random variables by mean values and variance defined by fuzzy measures, and the method is applicable to decision making with both randomness and fuzziness. Next, we compare several possible approaches regarding variances by examining them for some fuzzy random variables with values at triangle-type fuzzy numbers. We find the method with lambda-mean functions has proper properties, and we derive fundamental properties regarding the variance and the corresponding co-variance and correlation. Formulae are given to apply the results to triangle-type fuzzy numbers, trapezoidal-type fuzzy numbers, and some types of fuzzy random variables
{"title":"Mean Value and Variance of Fuzzy Random Variables by Evaluation Measures","authors":"Y. Yoshida","doi":"10.1109/IS.2006.348423","DOIUrl":"https://doi.org/10.1109/IS.2006.348423","url":null,"abstract":"This paper discusses an evaluation method of fuzzy numbers/fuzzy random variables by mean values and variance defined by fuzzy measures, and the method is applicable to decision making with both randomness and fuzziness. Next, we compare several possible approaches regarding variances by examining them for some fuzzy random variables with values at triangle-type fuzzy numbers. We find the method with lambda-mean functions has proper properties, and we derive fundamental properties regarding the variance and the corresponding co-variance and correlation. Formulae are given to apply the results to triangle-type fuzzy numbers, trapezoidal-type fuzzy numbers, and some types of fuzzy random variables","PeriodicalId":116809,"journal":{"name":"2006 3rd International IEEE Conference Intelligent Systems","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123280615","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 novel document clustering approach that solves some major problems of traditional document clustering approaches. Instead of depending on traditional vector space model, this approach represents documents as graphs using domain knowledge in ontology because graphs can represent the semantic relationships among the concepts in documents. Based on scale-free network theory, our approach generates a model for each document cluster from the ontology-enriched graph representation by identifying k high density subgraphs capturing the core semantic relationship information about each document cluster. Using these k high density subgraphs, each document is assigned to a proper document cluster. Our extensive experimental results on MEDLINE articles show that our approach outperforms two leading document clustering algorithms, BiSecting K-means and CLUTO's vcluster. Moreover, our approach provides a meaningful explanation for document clustering through generated models. This explanation helps users to understand clustering results and documents as a whole
{"title":"Clustering Ontology-enriched Graph Representation for Biomedical Documents based on Scale-Free Network Theory","authors":"Illhoi Yoo, Xiaohua Hu","doi":"10.1109/IS.2006.348532","DOIUrl":"https://doi.org/10.1109/IS.2006.348532","url":null,"abstract":"In this paper we introduce a novel document clustering approach that solves some major problems of traditional document clustering approaches. Instead of depending on traditional vector space model, this approach represents documents as graphs using domain knowledge in ontology because graphs can represent the semantic relationships among the concepts in documents. Based on scale-free network theory, our approach generates a model for each document cluster from the ontology-enriched graph representation by identifying k high density subgraphs capturing the core semantic relationship information about each document cluster. Using these k high density subgraphs, each document is assigned to a proper document cluster. Our extensive experimental results on MEDLINE articles show that our approach outperforms two leading document clustering algorithms, BiSecting K-means and CLUTO's vcluster. Moreover, our approach provides a meaningful explanation for document clustering through generated models. This explanation helps users to understand clustering results and documents as a whole","PeriodicalId":116809,"journal":{"name":"2006 3rd International IEEE Conference Intelligent Systems","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133007722","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}
Due to the interactions between the control channels, it is not an easy task to express the control strategies in the form of related multi-situations to multi-actions control fuzzy rules. Decoupled control is one answer to this problem. It separates the control task into two types: one is the dominating controller applied to fulfil the tracking task of a particular single-situation to a single-action loop, and the other is the compensator used to decouple the channels themselves. This paper adopts the self-organizing fuzzy logic control (SOFLC) strategy, which has the ability of self-generating and modifying the control rules depending on the on-line system control information, as the main controller for each channel. The compensating controller is triggered according to the nature of the effect of the interaction from the corresponding channel. The strategy of identifying the interaction effect follows the system performance evaluation method applied in SOFLC as well. A series of simulations were carried out on a two-input and two-output biomedical process, with the conclusion that the proposed decoupling control mechanism has the ability to deal with varying system dynamics, noise and inaccurate estimation of compensator gains very effectively
{"title":"Multivariable Self-organizing fuzzy logic control (SOFLC) using a switching mode linguistic compensator","authors":"Q. Lu, M. Mahfouf","doi":"10.1109/IS.2006.348425","DOIUrl":"https://doi.org/10.1109/IS.2006.348425","url":null,"abstract":"Due to the interactions between the control channels, it is not an easy task to express the control strategies in the form of related multi-situations to multi-actions control fuzzy rules. Decoupled control is one answer to this problem. It separates the control task into two types: one is the dominating controller applied to fulfil the tracking task of a particular single-situation to a single-action loop, and the other is the compensator used to decouple the channels themselves. This paper adopts the self-organizing fuzzy logic control (SOFLC) strategy, which has the ability of self-generating and modifying the control rules depending on the on-line system control information, as the main controller for each channel. The compensating controller is triggered according to the nature of the effect of the interaction from the corresponding channel. The strategy of identifying the interaction effect follows the system performance evaluation method applied in SOFLC as well. A series of simulations were carried out on a two-input and two-output biomedical process, with the conclusion that the proposed decoupling control mechanism has the ability to deal with varying system dynamics, noise and inaccurate estimation of compensator gains very effectively","PeriodicalId":116809,"journal":{"name":"2006 3rd International IEEE Conference Intelligent Systems","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125249717","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}
Most of the solutions proposed in image database applications are limited to a specific application domain. Generic models attempt to ease the development of applications to researchers. In this paper, to overcome the difficulties faced by application-specific systems, we present a general purpose image management model, oriented to fill the gap between systems and users. To the retrieval process the most important issue is to have a query model that efficiently represents the image nature integrated with traditional data and a feedback mechanism to model the user's information needs. This work develops a query language to deal with the fuzzy nature of images. The query language, I-OQL, based on the ODMG standard, also is able to define high level concepts and to integrate different levels of abstraction. We also propose a general-purpose relevance feedback mechanism oriented to fill the gap between systems and users, expressing user subjectivity in the retrieval process. Experiment results are presented to explore and validate the query refinement process
{"title":"A Query Model with Relevance Feedback for Image Database Retrieval","authors":"S. Montenegro Gonzalez, A. Yamakami","doi":"10.1109/IS.2006.348399","DOIUrl":"https://doi.org/10.1109/IS.2006.348399","url":null,"abstract":"Most of the solutions proposed in image database applications are limited to a specific application domain. Generic models attempt to ease the development of applications to researchers. In this paper, to overcome the difficulties faced by application-specific systems, we present a general purpose image management model, oriented to fill the gap between systems and users. To the retrieval process the most important issue is to have a query model that efficiently represents the image nature integrated with traditional data and a feedback mechanism to model the user's information needs. This work develops a query language to deal with the fuzzy nature of images. The query language, I-OQL, based on the ODMG standard, also is able to define high level concepts and to integrate different levels of abstraction. We also propose a general-purpose relevance feedback mechanism oriented to fill the gap between systems and users, expressing user subjectivity in the retrieval process. Experiment results are presented to explore and validate the query refinement process","PeriodicalId":116809,"journal":{"name":"2006 3rd International IEEE Conference Intelligent Systems","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127105097","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 paper built an analysis model of financial statements based on data mining methods, that is making data mining methods such as clustering, association rules and decision making tree work together to step by step go into deeper analysis of existing financial statements, during which a annual assets structure statement is worked out. The data used for research is from financial statements of electronic product corporations published on Internet. The paper established and implemented an integrated data mining model for the electronic product industry. Finally, some meaningful conclusions were drawn, which is great benefit to decision makers and investors in this industry to analyze financial situations of some corporate and make better investment decisions, budget or management plans
{"title":"An Analysis Model of Financial Statements Based on Data Mining","authors":"L. Yanhong, Liuyan Peng, Qin Zheng","doi":"10.1109/IS.2006.348531","DOIUrl":"https://doi.org/10.1109/IS.2006.348531","url":null,"abstract":"The paper built an analysis model of financial statements based on data mining methods, that is making data mining methods such as clustering, association rules and decision making tree work together to step by step go into deeper analysis of existing financial statements, during which a annual assets structure statement is worked out. The data used for research is from financial statements of electronic product corporations published on Internet. The paper established and implemented an integrated data mining model for the electronic product industry. Finally, some meaningful conclusions were drawn, which is great benefit to decision makers and investors in this industry to analyze financial situations of some corporate and make better investment decisions, budget or management plans","PeriodicalId":116809,"journal":{"name":"2006 3rd International IEEE Conference Intelligent Systems","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121994552","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：