遷移學(xué)習(xí)框架下不平衡分類問題研究
發(fā)布時(shí)間：2018-11-09 10:19
【摘要】：遷移學(xué)習(xí)是機(jī)器學(xué)習(xí)領(lǐng)域中新興的框架,放寬了傳統(tǒng)機(jī)器學(xué)習(xí)的兩個(gè)基本假設(shè),近年來(lái)受到了廣泛關(guān)注�，F(xiàn)有的關(guān)于遷移學(xué)習(xí)框架下不平衡分類問題的相關(guān)工作,主要集中于單源遷移研究方面,存在的潛在問題是可遷移信息較少,甚至可能會(huì)產(chǎn)生“負(fù)遷移”。針對(duì)已有遷移學(xué)習(xí)框架下不平衡分類問題相關(guān)研究存在的不足,本文通過引入多源遷移機(jī)制,展開了基于多源的遷移學(xué)習(xí)非均衡分類研究。首先,針對(duì)目標(biāo)領(lǐng)域和源領(lǐng)域數(shù)據(jù)分布相似且正負(fù)樣本不平衡的二分類遷移學(xué)習(xí)問題,論文提出一種基于多源數(shù)據(jù)的集成遷移學(xué)習(xí)非均衡樣本分類算法MSTUSC。該方法引入多個(gè)源領(lǐng)域數(shù)據(jù)以避免“負(fù)遷移”,采用新的樣本初始權(quán)重和樣本權(quán)重更新策略來(lái)解決不均衡樣本分類問題,并采用冗余樣本淘汰機(jī)制,適時(shí)淘汰多源域中冗余數(shù)據(jù),有效降低算法的時(shí)空開銷。在UCI標(biāo)準(zhǔn)數(shù)據(jù)上進(jìn)行實(shí)驗(yàn),采用F1值和AUC值作為評(píng)價(jià)指標(biāo)。實(shí)驗(yàn)結(jié)果表明,本文所提的MSTUSC算法在不平衡數(shù)據(jù)上的分類性能優(yōu)于其它幾種對(duì)比遷移算法。其次,為了改善MSTUSC算法的時(shí)間效率,還提出了面向分布式的多源數(shù)據(jù)的集成遷移學(xué)習(xí)非均衡樣本分類算法DMSTUSC。引入分布式系統(tǒng),將每個(gè)源領(lǐng)域劃分到分布式系統(tǒng)的一個(gè)節(jié)點(diǎn)上,在單個(gè)節(jié)點(diǎn)上進(jìn)行單源非均衡樣本分類的集成遷移學(xué)習(xí)算法訓(xùn)練,得到分類模型,最終將每個(gè)節(jié)點(diǎn)訓(xùn)練得到的分類模型進(jìn)行集成,得到多源數(shù)據(jù)的集成遷移學(xué)習(xí)非均衡樣本分類算法。通過實(shí)驗(yàn)分析可知,同MSTUSC算法相比,DMSTUSC算法的時(shí)間效率明顯提高。
[Abstract]:Transfer learning is a new framework in the field of machine learning. It has relaxed the two basic assumptions of traditional machine learning and has received extensive attention in recent years. The existing work on unbalanced classification in the framework of transfer learning is mainly focused on the study of single source migration. The potential problem is that there is less transferable information and even "negative migration" may occur. In view of the shortcomings of the existing researches on unbalanced classification under the framework of transfer learning, this paper introduces the mechanism of multi-source migration, and develops the research of non-equilibrium classification of transfer learning based on multi-source. First of all, aiming at the two-classification migration learning problem with similar data distribution in target domain and source domain and imbalance of positive and negative samples, this paper proposes an integrated migration learning disequilibrium sample classification algorithm MSTUSC. based on multi-source data. In this method, multiple source domain data are introduced to avoid "negative migration", new initial weight and weight updating strategies are adopted to solve the problem of uneven sample classification, and redundant sample elimination mechanism is adopted. Timely elimination of redundant data in multi-source domain can effectively reduce the space-time overhead of the algorithm. Based on the UCI standard data, F1 value and AUC value were used as the evaluation index. The experimental results show that the classification performance of the proposed MSTUSC algorithm on unbalanced data is better than that of other contrastive migration algorithms. Secondly, in order to improve the time efficiency of MSTUSC algorithm, a distributed multi-source data integration migration learning disequilibrium sample classification algorithm DMSTUSC. is proposed. The distributed system is introduced, each source domain is divided into one node of the distributed system, and the integrated migration learning algorithm of single source disequilibrium sample classification is trained on a single node, and the classification model is obtained. Finally, the classification model trained by each node is integrated, and an ensemble migration learning disequilibrium sample classification algorithm for multi-source data is obtained. The experimental results show that compared with the MSTUSC algorithm, the time efficiency of the DMSTUSC algorithm is obviously improved.
【學(xué)位授予單位】：安徽工業(yè)大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP181
本文編號(hào)：2320123
[Abstract]:Transfer learning is a new framework in the field of machine learning. It has relaxed the two basic assumptions of traditional machine learning and has received extensive attention in recent years. The existing work on unbalanced classification in the framework of transfer learning is mainly focused on the study of single source migration. The potential problem is that there is less transferable information and even "negative migration" may occur. In view of the shortcomings of the existing researches on unbalanced classification under the framework of transfer learning, this paper introduces the mechanism of multi-source migration, and develops the research of non-equilibrium classification of transfer learning based on multi-source. First of all, aiming at the two-classification migration learning problem with similar data distribution in target domain and source domain and imbalance of positive and negative samples, this paper proposes an integrated migration learning disequilibrium sample classification algorithm MSTUSC. based on multi-source data. In this method, multiple source domain data are introduced to avoid "negative migration", new initial weight and weight updating strategies are adopted to solve the problem of uneven sample classification, and redundant sample elimination mechanism is adopted. Timely elimination of redundant data in multi-source domain can effectively reduce the space-time overhead of the algorithm. Based on the UCI standard data, F1 value and AUC value were used as the evaluation index. The experimental results show that the classification performance of the proposed MSTUSC algorithm on unbalanced data is better than that of other contrastive migration algorithms. Secondly, in order to improve the time efficiency of MSTUSC algorithm, a distributed multi-source data integration migration learning disequilibrium sample classification algorithm DMSTUSC. is proposed. The distributed system is introduced, each source domain is divided into one node of the distributed system, and the integrated migration learning algorithm of single source disequilibrium sample classification is trained on a single node, and the classification model is obtained. Finally, the classification model trained by each node is integrated, and an ensemble migration learning disequilibrium sample classification algorithm for multi-source data is obtained. The experimental results show that compared with the MSTUSC algorithm, the time efficiency of the DMSTUSC algorithm is obviously improved.
【學(xué)位授予單位】：安徽工業(yè)大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP181
【參考文獻(xiàn)】
相關(guān)期刊論文 前7條
1 尹華;胡玉平;;一種代價(jià)敏感隨機(jī)森林算法[J];武漢大學(xué)學(xué)報(bào)(工學(xué)版);2014年05期
2 顧鑫;王士同;;大樣本多源域與小目標(biāo)域的跨領(lǐng)域快速分類學(xué)習(xí)[J];計(jì)算機(jī)研究與發(fā)展;2014年03期
3 張倩;李明;王雪松;程玉虎;朱美強(qiáng);;一種面向多源領(lǐng)域的實(shí)例遷移學(xué)習(xí)[J];自動(dòng)化學(xué)報(bào);2014年06期
4 于重重;田蕊;譚勵(lì);涂序彥;;非平衡樣本分類的集成遷移學(xué)習(xí)算法[J];電子學(xué)報(bào);2012年07期
5 趙秀寬;陽(yáng)建宏;黎敏;徐金梧;;一種改進(jìn)的不平衡數(shù)據(jù)集分類方法[J];計(jì)算機(jī)工程;2011年15期
6 歐陽(yáng)震諍;羅建書;胡東敏;吳泉源;;一種不平衡數(shù)據(jù)流集成分類模型[J];電子學(xué)報(bào);2010年01期
7 王和勇;樊泓坤;姚正安;李成安;;不平衡數(shù)據(jù)集的分類方法研究[J];計(jì)算機(jī)應(yīng)用研究;2008年05期
，本文編號(hào)：2320123
本文鏈接：http://sikaile.net/kejilunwen/zidonghuakongzhilunwen/2320123.html
最近更新
教材專著
