本文關鍵詞： 半監(jiān)督分類 流形學習 自步學習 稀疏編碼 半監(jiān)督學習　出處：《首都經(jīng)濟貿(mào)易大學》2017年碩士論文　論文類型：學位論文

【摘要】：對數(shù)據(jù)進行標注是一項冗繁、耗時且容易引起誤差的工作,一方面這使得有標簽數(shù)據(jù)的數(shù)目比較有限,而相對地無標簽數(shù)據(jù)的數(shù)目則比較多;另一方面,在對數(shù)據(jù)進行標注過程中,數(shù)據(jù)標簽容易出現(xiàn)錯誤,而很多機器學習的算法對標簽帶噪聲的數(shù)據(jù)是敏感的。因此,人們希望探究出能利用大量無標簽數(shù)據(jù)且能對噪聲標簽數(shù)據(jù)比較穩(wěn)健的算法。針對此問題,本文提出了一種稱為自步流形正則化的框架,它能利用少量有標簽數(shù)據(jù)和大量無標簽數(shù)據(jù)進行半監(jiān)督分類,且對含噪聲標簽的數(shù)據(jù)具有較好的穩(wěn)健性。具體地,該方法在半監(jiān)督分類的流形正則化框架中引入自步學習正則化項,以此來穩(wěn)健地選取有標簽的訓練數(shù)據(jù);同時,我們利用數(shù)據(jù)稀疏編碼中蘊含的判別信息來控制分類器的光滑度,提高分類性能。最后,我們針對構(gòu)造的優(yōu)化問題設計一種交替搜索策略,得到具有顯性表達式的分類函數(shù)。該方法可適用于多分類問題,在對具噪聲標簽的數(shù)據(jù)保持穩(wěn)健性的同時兼顧了分類器的復雜度和光滑度,使得模型有較小的泛化誤差。三個數(shù)據(jù)集上的實驗結(jié)果顯示出此方法在含有噪聲標簽數(shù)據(jù)的分類效果上優(yōu)于傳統(tǒng)半監(jiān)督分類方法。
[Abstract]:Tagging data is a redundant, time-consuming and error-prone work. On the one hand, the number of labeled data is limited, while the number of untagged data is relatively large. On the other hand, in the process of tagging data, data labels are prone to errors, and many machine learning algorithms are sensitive to label noisy data. People want to explore an algorithm that can use a lot of untagged data and be more robust to noise tagged data. To solve this problem, a framework called self-stepping manifold regularization is proposed in this paper. It can use a small amount of labeled data and a large number of untagged data for semi-supervised classification, and it has good robustness to the data with noise labels. In this method, self-learning regularization items are introduced into the manifold regularization framework of semi-supervised classification, so that the tagged training data can be selected stably. At the same time, we use the discriminant information contained in the sparse data coding to control the smoothness of the classifier and improve the classification performance. Finally, we design an alternative search strategy for the structural optimization problem. The classification function with explicit expression is obtained. This method can be applied to multi-classification problems. It keeps the robustness of the noisy data and takes into account the complexity and smoothness of the classifier. The experimental results on three data sets show that the proposed method is better than the traditional semi-supervised classification method in classification with noisy label data.
【學位授予單位】：首都經(jīng)濟貿(mào)易大學
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：TP18

【參考文獻】

相關期刊論文 前2條

1 張晨光;李玉擰;;哈希圖半監(jiān)督學習方法及其在圖像分割中的應用[J];自動化學報;2010年11期

2 張晨光;李玉擰;;基于半監(jiān)督學習的眉毛圖像分割方法[J];計算機工程與應用;2009年21期

，

本文編號：1465955