本文選題：多標(biāo)簽分類 + 在線學(xué)習(xí)�。� 參考：《南京師范大學(xué)》2017年碩士論文

【摘要】：在多標(biāo)簽分類問題中,一個(gè)樣本可以同時(shí)屬于多個(gè)類別標(biāo)簽,且樣本標(biāo)簽之間不再相互排斥。目前,多標(biāo)簽分類問題已在文本分類,自然場景分類和音樂情感標(biāo)注等領(lǐng)域得到廣泛應(yīng)用,因此提出很多多標(biāo)簽分類算法。近年來,多標(biāo)簽分類算法大多采用批量學(xué)習(xí)的方式,它要求將整個(gè)訓(xùn)練數(shù)據(jù)集全部讀入內(nèi)存且可以通過一次學(xué)習(xí)得到最終分類模型。但在實(shí)際應(yīng)用中,尤其對(duì)大規(guī)模數(shù)據(jù)集的分類問題,這種批量學(xué)習(xí)的方式,將會(huì)消耗大量的時(shí)間和空間資源。針對(duì)上述問題,本文基于在線學(xué)習(xí)理論,圍繞大規(guī)模多標(biāo)簽分類問題展開研究,提出了兩種多標(biāo)簽在線分類算法。主要工作如下:1.使用二類相關(guān)分解策略,結(jié)合已有的二類在線“被動(dòng)-進(jìn)攻”主動(dòng)學(xué)習(xí)算法,提出基于分解策略的多標(biāo)簽在線“被動(dòng)-進(jìn)攻”主動(dòng)學(xué)習(xí)算法(MLPAA)。算法采用主動(dòng)學(xué)習(xí)的方式查詢多標(biāo)簽樣本信息,這樣不僅可以利用在線學(xué)習(xí)的方式不斷更新多標(biāo)簽分類器模型,還利用主動(dòng)學(xué)習(xí)的方式探索未標(biāo)注樣本信息,減少人工標(biāo)注代價(jià)和時(shí)間。在實(shí)驗(yàn)中,根據(jù)五個(gè)多標(biāo)簽評(píng)價(jià)準(zhǔn)則,在八個(gè)多標(biāo)簽數(shù)據(jù)集上,將MLPAA算法與三個(gè)算法進(jìn)行實(shí)驗(yàn)比較。結(jié)果表明,MLPAA算法相對(duì)于MLRPE, MLPEA和MLRPA算法具有更好的分類性能。2.基于標(biāo)簽排序思想,改進(jìn)多類在線“被動(dòng)-進(jìn)攻”分類算法,提出了考慮標(biāo)簽相關(guān)性的多標(biāo)簽在線“被動(dòng)-進(jìn)攻”分類算法(MLLRPA)。算法通過最大化多標(biāo)簽樣本中相關(guān)標(biāo)簽子集與不相關(guān)標(biāo)簽子集之間的間隔,預(yù)測標(biāo)簽排序?qū)Φ姆绞浇⑾嚓P(guān)標(biāo)簽與不相關(guān)標(biāo)簽的排序錯(cuò)誤集合,根據(jù)錯(cuò)誤集合的大小,進(jìn)而更新分類器模型。在實(shí)驗(yàn)中,在10個(gè)多標(biāo)簽基準(zhǔn)數(shù)據(jù)集上,根據(jù)四個(gè)多標(biāo)簽評(píng)價(jià)指標(biāo),將MLLRPA算法與在線MMP, BR-PE和BR-PA三個(gè)算法進(jìn)行對(duì)比實(shí)驗(yàn)。實(shí)驗(yàn)結(jié)果表明本文提出的MLLRPA算法具有較好的性能。
[Abstract]:In the multi-label classification problem, a sample can belong to multiple class labels at the same time, and the sample labels are no longer mutually exclusive. At present, multi-label classification problem has been widely used in the fields of text classification, natural scene classification and music emotion tagging, so many multi-label classification algorithms are proposed. In recent years, most multi-label classification algorithms adopt batch learning, which requires the entire training data set to be read into memory and the final classification model can be obtained by one learning. However, in practical applications, especially for the classification of large-scale data sets, this mass learning method will consume a lot of time and space resources. In order to solve the above problems, based on the online learning theory, this paper focuses on the large-scale multi-label classification problem, and proposes two online multi-label classification algorithms. The main work is as follows: 1. Based on the two-class correlation decomposition strategy and the existing passive attack active learning algorithm, a multi-label online passive attack active learning algorithm based on decomposition strategy is proposed. The algorithm uses active learning to query multi-label sample information, which can not only update the multi-label classifier model by online learning, but also explore the unlabeled sample information by active learning. Reduce manual marking cost and time. In the experiment, according to five multi-label evaluation criteria, the MLPAA algorithm is compared with three algorithms on eight multi-label datasets. The results show that the MLPAA algorithm has better classification performance than MLRPE, MLPEA and MLRPA algorithms. Based on the idea of label sorting, a multi-class online "passive-attack" classification algorithm considering label correlation is proposed, and a multi-label online "passive-attack" classification algorithm is proposed. By maximizing the interval between the correlation tag subset and the unrelated tag subset in the multi-label sample, the algorithm establishes the sorting error set of the correlation label and the unrelated label by predicting the sorting pairs of the label, according to the size of the error set. Then the classifier model is updated. In the experiment, the MLLRPA algorithm is compared with the online MMP, BR-PE and BR-PA algorithms on 10 multi-label datum data sets according to four multi-label evaluation indexes. Experimental results show that the proposed MLLRPA algorithm has better performance.
【學(xué)位授予單位】：南京師范大學(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1條

1 徐美香;孫福明;李豪杰;;主動(dòng)學(xué)習(xí)的多標(biāo)簽圖像在線分類[J];中國圖象圖形學(xué)報(bào);2015年02期

，

本文編號(hào)：1800907