DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Kil, Rhee-Man | - |
dc.contributor.advisor | 길이만 | - |
dc.contributor.advisor | Kim, Sung-Ho | - |
dc.contributor.advisor | 김성호 | - |
dc.contributor.author | Rosas, Harvey | - |
dc.contributor.author | 로사스 하비 | - |
dc.date.accessioned | 2013-09-12T02:32:25Z | - |
dc.date.available | 2013-09-12T02:32:25Z | - |
dc.date.issued | 2011 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=482596&flag=dissertation | - |
dc.identifier.uri | http://hdl.handle.net/10203/181563 | - |
dc.description | 학위논문(박사) - 한국과학기술원 : 수리과학과, 2011.8, [ v, 45 p. ] | - |
dc.description.abstract | Multi-labeled classification presents a challenging problem in data mining. Furthermore, it has be- come a very important research field due to the need of handling large scale databases, with some of them having an incredible amount of information in text format. Thus, ecient methods and automatic tools had recently gained relevance. This work focuses on the efforts to improve the performance of the automatic text multi-categorical multi-labeled classification using SVM classifiers. We introduce a new method of multi-labeled classification based on a class probability output network called 2 layer Conditional Probability Output Networks. With the objective of refining the classification accuracy, the output of the support vector machine is considered in order to get a complete distribution inde- pendent algorithm, both kernel and probability distribution parameters are finely tuned to improve its performance, furthermore a new method for multi-labeled classification based on a complete distribution and an uncertainty measure is proposed. Experiments are done using 2 different data frameworks for classification problems: multimedia data filtering and Reuters-21578 modapte as benchmark data-sets, the effectiveness of the method is compared in terms accuracy and micro and macro averaging F1- mea- sure. | eng |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | Data Mining | - |
dc.subject | Text Mining | - |
dc.subject | Multilabeled Text Classification | - |
dc.subject | 데이터 마이닝 | - |
dc.subject | 텍스트 마이닝 | - |
dc.subject | 자동 텍스트 구분 | - |
dc.subject | 다중 텍스트 라벨 분류 | - |
dc.subject | Automatic Text Classification | - |
dc.title | Text mining based on conditional probability output networks | - |
dc.title.alternative | 조건부 확률망에 기초한 텍스트 마이닝 | - |
dc.type | Thesis(Ph.D) | - |
dc.identifier.CNRN | 482596/325007 | - |
dc.description.department | 한국과학기술원 : 수리과학과, | - |
dc.identifier.uid | 020064515 | - |
dc.contributor.localauthor | Kil, Rhee-Man | - |
dc.contributor.localauthor | 길이만 | - |
dc.contributor.localauthor | Kim, Sung-Ho | - |
dc.contributor.localauthor | 김성호 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.