检索结果-内蒙古大学图书馆

Support vector regression based hybrid rule extraction methods for forecasting

EXPERT SYSTEMS WITH APPLICATIONS 2010年第8期37卷 5577-5589页

作者： Farquad, M. A. H. Ravi, V. Raju, S. Bapi Inst Dev & Res Banking Technol Hyderabad 500057 Andhra Pradesh India Univ Hyderabad Dept Comp & Informat Sci Hyderabad 500046 Andhra Pradesh India

Support Vector regression (SVR) solves regression problems based on the concept of Support Vector Machine (SVM) introduced by Vapnik (1995). The main drawback of these newer techniques is their lack of interpretability. In other words, it is difficult for the human analyst to understand the knowledge learnt by these models during training. The most popular way to overcome this difficulty is to extract if-then rules from SVM and SVR. Rules provide explanation capability to these models and improve the comprehensibility of the system. Over the last decade, different algorithms for extracting rules from SVM have been developed. However rule extraction from SVR is not widely available yet. In this paper a novel hybrid approach for extracting rules from SVR is presented. The proposed hybrid rule extraction procedure has two phases: (1) Obtain the reduced training set in the form of support vectors using SVR (2) Train the machine leaning techniques (with explanation capability) using the reduced training set. Machine learning techniques viz., classification and regression tree (CART), Adaptive Network based Fuzzy Inference System (ANFIS) and Dynamic Evolving Fuzzy Inference System (DENFIS) are used in the phase 2. The proposed hybrid rule extraction procedure is compared to stand-alone CART. ANFIS and DENFIS. Extensive experiments are conducted on five benchmark data sets viz. Auto MPG, Body Fat, Boston Housing, Forest Fires and Pollution, to demonstrate the effectiveness of the proposed approach in generating accurate regression rules. The efficiency of these techniques is measured using Root Mean Squared Error (RMSE). From the results obtained, it is concluded that when the support vectors with the corresponding predicted target values are used, the SVR based hybrids outperform the stand-alone intelligent techniques and also the case when the support vectors with the corresponding actual target values are used. (c) 2010 Elsevier Ltd. All rights reserved.

关键词： Rule extraction Support vector machine Support vector regression classification and regression tree Adaptive network based fuzzy inference system Dynamic evolving fuzzy inference system Root mean squared error

来源：评论

学校读者我要写书评

暂无评论

Low Flows Regionalization in North-Western Italy

引用

WATER RESOURCES MANAGEMENT 2010年第14期24卷 4049-4074页

作者： Vezza, Paolo Comoglio, Claudio Rosso, Maurizio Viglione, Alberto Politecn Torino Dipartimento Idraul Trasporti & Infrastrutture I-10129 Turin Italy Politecn Torino Dipartimento Territorio Ambiente & Geotecnol I-10129 Turin Italy Tech Univ Wien Inst Wasserbau & Ingn Hydrol A-1040 Vienna Austria

Prediction of low flows in ungauged catchments is needed in many branches of water resources management, including water availability and river ecology studies. In this paper we analyze the regional variability of q (95), i.e., the specific discharge that is exceeded 95% of the time, in North-Western Italy (Piemonte and Valle d'Aosta Regions). Multiple regressions with morphoclimatic catchment characteristics are applied in subregions obtained through four classification methods: Seasonality Indices (SI), classification and regression trees (CRT), Residual Pattern Approach (RPA) and Weighted Cluster Analysis (WCA). All the classification methods separate the South-Eastern Apennine-Mediterranean area from the rest of the study domain (the Alps mountain range), even if they use different criteria to carry out this division (e.g., the percentage of forest, seasonality of low flows, combination of several parameters). In the Apennine-Mediterranean part of the area, low flows occur in summer with a long period of drought and are mainly due to dry climate, moderate snowpack storage and high evapotranspiration. In Alpine catchments low flows occur in winter and vary according to precipitation, elevation, interactions with aquifers and land cover. Within the Alpine mountain range the CRT algorithm identifies a number of small high-elevation catchments in which the intense drought period during winter has the soil freezing processes as the driving force. From a statistical point of view, the CRT model outperforms the models obtained by the other classification techniques in terms of explained variance (69%). Because of this, and given the meaningful hydrological interpretation of the results, we use the CRT model for the regionalization of q (95) in Piemonte and Valle d'Aosta. Lastly, as operational procedure for future low flow regionalization studies, we suggest that more classification methods should be applied to assist the critical analysis of the results.

关键词： Low flows Regionalization Regional regression classification and regression tree CART

来源：评论

学校读者我要写书评

暂无评论

Methodology for assessing determinants of manure use in urban areas of Africa

引用

WASTE MANAGEMENT & RESEARCH 2010年第12期28卷 1076-1086页

作者： Suman, Paul Thys, Eric Mfoukou-Ntsakala, Andre Ali, Laouali Ouedraogo, Mamadou Van den Bossche, Peter Van Huylenbroeck, Guido Berkvens, Dirk Speybroeck, Niko Inst Trop Med Natl Str 150 B-2000 Antwerp Belgium Sylhet Agr Univ Sylhet Bangladesh Ctr Rech Vet & Zootech Brazzaville Rep Congo Direct Reg Ressources Anim Maradi Niger Projet PCZLD PATTEC BF Bobo Dioulasso Burkina Faso Fac Biosci Ghent Belgium

The objectives of this study were to identify the major determinants (e.g. zootechnical and socio-economic parameters) of manure management practised by urban livestock keepers. Data were collected from four single-visit, multi-object surveys in three cities of Africa. The livestock keepers were classified either as 'good user' or 'bad user' of manure depending on how they use the manure. The results of this study indicate that the habit of manure utilization shows geographical differences. In general, livestock keepers engaged in agricultural activities are good manure users. Furthermore, manual workers, traders or the self-employed are better manure users than other professional groups. The manure utilization varies also in function of the livestock numbers. In Brazzaville, the habit of selling animals influenced manure management positively. Experience of keeping livestock before the civil war in Brazzaville was also a determining factor of manure use in the post-war urban livestock keeping in 2001. When designing policies to reduce the environmental effects of urban livestock keeping, the highlighted factors can assist city planners to understand how urban livestock keepers use manure. Using the non-parametric classification tree method CART, two approaches were used to select an optimal classification of 'bad' manure users: the 'Cook and Goldman rule' and the 'classification with different misclassification cost ratios' method. The optimal tree was selected by using receiver operating characteristics curve information. This method can be useful in assessing other aspects of waste management.

关键词： classification and regression tree cost ratio misclassification receiver operating characteristic curve manure socio-economic status

来源：评论

学校读者我要写书评

暂无评论

DATA MINING FOR OCCUPATIONAL INJURY RISK: A CASE STUDY

引用

INTERNATIONAL JOURNAL OF RELIABILITY QUALITY & SAFETY ENGINEERING 2010年第4期17卷 351-380页

作者： Bevilacqua, Maurizio Ciarapica, Filippo Emanuele Giacchetta, Giancarlo Univ Politecn Marche Dept Energy Via Brecce Bianche I-60131 Ancona Italy

The management of occupational injury is of strategic importance in a company from the organizational, engineering and economic point of view. This work is an attempt to apply data mining techniques to data regarding accidents in a medium-sized refinery. Several techniques were adopted in order to identify the important relationships between risk level and immediate/root causes and corrective actions. As a Data Mining technique were tested: Negative Binomial regression (NBR), Chi-Squared Automatic Interaction Detection (CHAID);Exhaustive CHAID;classification and regression trees (CART);Quick, Unbiased, Efficient Statistical tree (QUEST), Artificial Neural Network (ANN) and Neuro-Fuzzy Systems (FIS). The comparison carried out in this study shows through a real application the flexibility and advantages of using the neuro-fuzzy network, a typical soft computing tool. Using these innovative techniques to analyse injury data this study aims to: obtain a classification of input data according to their importance and/or influence on the risk level in injuries;assess how a variation in one or more pieces of input data can effect occupational injury and subsequently carry out a sensitivity analysis concerning the probability, the consequences and risks of the injurie events;The analyses carried out indicated important relationships between the variables, providing useful decision-making rules which can be followed when adopting measures for improvement.

关键词： Occupational injury data mining classification and regression tree neural network neuro-fuzzy techniques

来源：评论

学校读者我要写书评

暂无评论

Pharmacogenetics of tomorrow: the 1+1=3 principle

引用

PHARMACOGENOMICS 2010年第7期11卷 1011-1017页

作者： Pander, Jan Wessels, Judith A. M. Mathijssen, Ron H. J. Gelderblom, Hans Guchelaar, Henk Jan Leiden Univ Ctr Med Dept Clin Pharm & Toxicol NL-2300 RC Leiden Netherlands Dr Daniel Den Hoed Canc Ctr Erasmus Med Ctr Dept Med Oncol NL-3008 AE Rotterdam Netherlands Leiden Univ Ctr Med Dept Clin Oncol NL-2300 RC Leiden Netherlands

Disappointing results from replicating pharmacogenetic association studies have prompted the search for novel statistical techniques to analyze the data, while taking into account the biological complexity underlying drug response. Two of these techniques multifactor dimensionality reduction and classification and regression tree analysis will probably be applied in increasing numbers of future pharmacogenetic studies. In this article, we describe the concepts underlying both techniques and illustrate their application in a recent pharmacogenetic study.

关键词： classification and regression tree multifactor dimensionality reduction pharmacogenetics sunitinib toxicity

来源：评论

学校读者我要写书评

暂无评论

Process disturbance identification through integration of spatiotemporal ICA and CART approach

引用

NEURAL COMPUTING & APPLICATIONS 2010年第5期19卷 677-689页

作者： Chiu, Chih-Chou Hwang, Shin-Ying Cook, Deborah F. Luh, Yuan-Ping Natl Taipei Univ Technol Inst Commerce Automat & Management Taipei Taiwan Natl Chengchi Univ Dept Management Informat Syst Taipei 11623 Taiwan Virginia Polytech Inst & State Univ Dept Business Informat Technol Blacksburg VA 24061 USA Natl Taipei Univ Technol Dept Mech Engn Taipei Taiwan

Many studies have been conducted over the past several years evaluating the integrated use of statistical process control (SPC) and engineering process control (EPC). The majority of these studies reported that combining SPC with EPC outperforms the use of only SPC or EPC. Basically, the former aims to rapidly detect assignable causes and time points for abnormalities that take place during process;and the latter is a method in which input variables are adjusted against process outputs through a feedback control mechanism. Although combining SPC with EPC can effectively detect time points when abnormalities occur during process, their combination can also cause an increased occurrence of false alarms when autocorrelation is present in the process. In this study, to increase the accuracy of process disturbance identification, we propose the integration of spatiotemporal independent component analysis (stICA) with the classification and regression tree (CART) approach to improve our capability to identify process disturbances and recognize shifts in the correlated process parameters. The integration of the CART methodology results in the development of decision rules that can provide valuable information related to the impact of variation in process variable values. These decision rules can provide an increased understanding of process behavior and useful information for process control. For comparison, the integration of traditional principle component analysis (PCA) with CART (called PCA-CART), ICA with CART (called ICA-CART) and cumulative sum chart approaches were applied to evaluate the identification capability of the proposed approach. As the results reveal, the proposed approach is more effective for monitoring correlated process.

关键词： Process disturbance Statistical process control Engineering process control Spatiotemporal independent component analysis classification and regression tree

来源：评论

学校读者我要写书评

暂无评论

Improving marketing response by data mining in social network

Improving marketing response by data mining in social networ...

引用

International Conference on Advances in Social Network Analysis and Mining (ASONAM)

作者： Surma, Jerzy Furmanek, Anna Warsaw Sch Econ Coll Business Adm Warsaw Poland Warsaw Sch Econ Coll Econ Anal Warsaw Poland

ISBN: (纸本)9780769541389

Social networks have generated great expectations connected with their potential business value. The purpose of our research is to present that even a rudimentary application of data mining techniques can bring statistically significant improvement in marketing response accuracy throughout the virtual community. In our test the C&RT (classification and regression tree) approach was used to generate a classification tree that allows us to formulate some specific rules to identify the proper target group. In the performed empirical experiments, based on the real social network data, we showed that it is possible to improve marketing response. This promising result was obtained without any advanced and time consuming transformation of the available data.

关键词： Internet classification and regression tree classification tree data mining data mining data mining marketing data processing marketing response marketing response improvement regression analysis social network social network social networking (online) tree data structures virtual community

来源：评论

学校读者我要写书评

暂无评论

A hybrid Decision Support System for the Risk Assessment of retinopathy development as a long term complication of Type 1 Diabetes Mellitus

A hybrid Decision Support System for the Risk Assessment of ...

引用

32nd Annual International Conference of the IEEE Engineering-in-Medicine-and-Biology-Society (EMBC 10)

作者： Skevofilakas, Marios Zarkogianni, Konstantia Karamanos, Basil G. Nikita, Konstantina S. Natl Tech Univ Athens Sch Elect & Comp Engn 9 Heroon Polytech Str GR-15780 Athens Greece Hippokrateion Hosp Diabetes Ctr GR-11527 Athens Greece

ISBN: (纸本)9781424441242

The aim of the present study is to design and develop a Decision Support System (DSS) closely coupled with an Electronic Medical Record (EMR), able to predict the risk of a Type 1 Diabetes Mellitus (T1DM) patient to develop retinopathy. The proposed system is able to store a wealth of information regarding the clinical state of the T1DM patient and continuously provide the health experts with predictions regarding the possible future complications that he may present. The DSS is a hybrid infrastructure combining a Feedforward Neural Network (FNN), a classification and regression tree (CART) and a Rule Induction C5.0 classifier, with an improved Hybrid Wavelet Neural Network (iHWNN). A voting mechanism is utilized to merge the results from the four classification models. The proposed DSS has been trained and evaluated using data from 55 T1DM patients, acquired by the Athens Hippokration Hospital in close collaboration with the EURODIAB research team. The DSS has shown an excellent performance resulting in an accuracy of 98%. Care has been taken to design and implement a consistent and continuously evolving Information Technology (IT) system by utilizing technologies such as smart agents periodically triggered to retrain the DSS with new cases added in the data repository.

关键词： CART DSS EMR EURODIAB FNN classification and regression tree decision support systems diseases electronic medical record eye feedforward neural nets feedforward neural network hybrid decision support system hybrid wavelet neural network iHWNN medical computing medical information systems retinopathy risk assessment type 1 diabetes mellitus wavelet transforms

来源：评论

学校读者我要写书评

暂无评论

Evidence on women trafficked for sexual exploitation: A rights based analysis

引用

EUROPEAN JOURNAL OF LAW AND ECONOMICS 2010年第1期29卷 15-42页

作者： Bettio, Francesca Nandi, Tushar K. Univ Siena Dept Econ I-53100 Siena Italy

The aim of this paper is to investigate which factors influence the pattern of enforcement (violation) of basic rights among women trafficked for sexual exploitation. A conceptual framework is adopted where the degree of agency and the possibility to influence the terms of sex-based transactions are seen as conditional on the enforcement of some basic rights. Using data collected by the International Organization for Migration (IOM) on women assisted by the organization after having been trafficked for sexual exploitation, we investigate the enforcement (violation) of five uncompromisable rights, namely the right to physical integrity, to move freely, to have access to medical care, to use condoms, and to exercise choice over sexual services. By combining classification tree analysis and ordered probit estimation we find that working location and country of work are the main determinants of rights enforcement, while individual and family characteristics play a marginal role. Specifically, we find that (1) in lower market segments working on the street is comparatively less 'at risk' of rights violation;(2) there is no consistently 'good' or 'bad' country of work, but public awareness on trafficking within the country is important;(3) the strength of organised crime in the country of work matters only in conjunction with other local factors, and (4) being trafficked within one's country, as opposed to being trafficked internationally, is associated with higher risk of rights violation.

关键词： Human trafficking Sexual exploitation Basic rights classification and regression tree Ordered probit

来源：评论

学校读者我要写书评

暂无评论

Identifying type 1 and type 2 diabetic cases using administrative data: a tree-structured model

引用

Journal of diabetes science and technology 2011年第3期5卷 486-93页

作者： Weihsuan Lo-Ciganic Janice C Zgibor Kristine Ruppert Vincent C Arena Roslyn A Stone Department of Epidemiology Graduate School of Public Health University of Pittsburgh Pittsburgh Pennsylvania USA. wel32@pitt.edu

BACKGROUND:To date, few administrative diabetes mellitus (DM) registries have distinguished type 1 diabetes mellitus (T1DM) from type 2 diabetes mellitus (T2DM). OBJECTIVE:Using a classification tree model, a prediction rule was developed to distinguish T1DM from T2DM in a large administrative database. METHODS:The Medical Archival Retrieval System at the University of Pittsburgh Medical Center included administrative and clinical data from January 1, 2000, through September 30, 2009, for 209,647 DM patients aged ≥18 years. Probable cases (8,173 T1DM and 125,111 T2DM) were identified by applying clinical criteria to administrative data. Nonparametric classification tree models were fit using TIBCO Spotfire S+ 8.1 (TIBCO Software), with model size based on 10-fold cross validation. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of T1DM were estimated. RESULTS:The main predictors that distinguished T1DM from T2DM are age <40 years; International classification of Disease, 9th revision, codes of T1DM or T2DM diagnosis; inpatient oral hypoglycemic agent use; inpatient insulin use; and episode(s) of diabetic ketoacidosis diagnosis. Compared with a complex clinical algorithm, the tree-structured model to predict T1DM had 92.8% sensitivity, 99.3% specificity, 89.5% PPV, and 99.5% NPV. CONCLUSION:The preliminary predictive rule appears to be promising. Being able to distinguish between DM subtypes in administrative databases will allow large-scale subtype-specific analyses of medical care costs, morbidity, and mortality.

关键词： administrative database classification and regression tree diabetes Medical Archival Retrieval System negative predictive value positive predictive value sensitivity specificity

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：