Active Learning for Multiclass Cost-sensitive Classification Using Probabilistic Models

碩士 === 國立臺灣大學 === 資訊工程學研究所 === 100 === Multiclass cost-sensitive active learning is a relatively new problem. In this thesis, we derive the maximum expected cost and cost-weighted minimum margin strategy for multiclass cost-sensitive active learning. These two strategies can be seem as the extended...

Full description

Bibliographic Details
Main Authors:	Po-Lung Chen, 陳柏龍
Other Authors:	Hsuan-Tien Lin
Format:	Others
Language:	en_US
Published:	2012
Online Access:	http://ndltd.ncl.edu.tw/handle/65244803215661729379

id	ndltd-TW-100NTU05392132
record_format	oai_dc
spelling	ndltd-TW-100NTU053921322015-10-13T21:50:44Z http://ndltd.ncl.edu.tw/handle/65244803215661729379 Active Learning for Multiclass Cost-sensitive Classification Using Probabilistic Models 使用機率模型實行成本導向多重分類的主動學習演算法 Po-Lung Chen 陳柏龍碩士國立臺灣大學資訊工程學研究所 100 Multiclass cost-sensitive active learning is a relatively new problem. In this thesis, we derive the maximum expected cost and cost-weighted minimum margin strategy for multiclass cost-sensitive active learning. These two strategies can be seem as the extended version of classical cost-insensitive active learning strategies. The experimental results demonstrate that the derived strategies are promising for cost-sensitive active learning. In particular, the cost-sensitive strategies outperform cost-insensitive ones on many benchmark data sets. The results also reveal how the hardness of data affects the performance of active learning strategies. Thus, in practical active learning applications, data analysis before strategy selection can be important. Hsuan-Tien Lin 林軒田 2012 學位論文 ; thesis 38 en_US
collection	NDLTD
language	en_US
format	Others
sources	NDLTD
description	碩士 === 國立臺灣大學 === 資訊工程學研究所 === 100 === Multiclass cost-sensitive active learning is a relatively new problem. In this thesis, we derive the maximum expected cost and cost-weighted minimum margin strategy for multiclass cost-sensitive active learning. These two strategies can be seem as the extended version of classical cost-insensitive active learning strategies. The experimental results demonstrate that the derived strategies are promising for cost-sensitive active learning. In particular, the cost-sensitive strategies outperform cost-insensitive ones on many benchmark data sets. The results also reveal how the hardness of data affects the performance of active learning strategies. Thus, in practical active learning applications, data analysis before strategy selection can be important.
author2	Hsuan-Tien Lin
author_facet	Hsuan-Tien Lin Po-Lung Chen 陳柏龍
author	Po-Lung Chen 陳柏龍
spellingShingle	Po-Lung Chen 陳柏龍 Active Learning for Multiclass Cost-sensitive Classification Using Probabilistic Models
author_sort	Po-Lung Chen
title	Active Learning for Multiclass Cost-sensitive Classification Using Probabilistic Models
title_short	Active Learning for Multiclass Cost-sensitive Classification Using Probabilistic Models
title_full	Active Learning for Multiclass Cost-sensitive Classification Using Probabilistic Models
title_fullStr	Active Learning for Multiclass Cost-sensitive Classification Using Probabilistic Models
title_full_unstemmed	Active Learning for Multiclass Cost-sensitive Classification Using Probabilistic Models
title_sort	active learning for multiclass cost-sensitive classification using probabilistic models
publishDate	2012
url	http://ndltd.ncl.edu.tw/handle/65244803215661729379
work_keys_str_mv	AT polungchen activelearningformulticlasscostsensitiveclassificationusingprobabilisticmodels AT chénbǎilóng activelearningformulticlasscostsensitiveclassificationusingprobabilisticmodels AT polungchen shǐyòngjīlǜmóxíngshíxíngchéngběndǎoxiàngduōzhòngfēnlèidezhǔdòngxuéxíyǎnsuànfǎ AT chénbǎilóng shǐyòngjīlǜmóxíngshíxíngchéngběndǎoxiàngduōzhòngfēnlèidezhǔdòngxuéxíyǎnsuànfǎ
_version_	1718069400375918592

Active Learning for Multiclass Cost-sensitive Classification Using Probabilistic Models

Similar Items