本文選題：計算機　切入點：算法　出處：《山東師范大學(xué)》2017年碩士論文　論文類型：學(xué)位論文

【摘要】：在處理計算機問題時,現(xiàn)實中遇到的數(shù)據(jù)一般都是高維度的,存在很多不相關(guān)的冗長特征。這為現(xiàn)實問題的解決帶來了一定的困難。人們研究出了特征選擇算法,以此提高算法選擇的準(zhǔn)確率。此方法可以高效的對數(shù)據(jù)進行降維,能夠從數(shù)據(jù)的原始特征中直接選擇出最優(yōu)化的特征子集。因此,針對這一課題的研究已成為機器學(xué)習(xí)和數(shù)據(jù)挖掘領(lǐng)域的熱點研究課題。在解決實際問題時也會發(fā)現(xiàn)數(shù)據(jù)之間具有多個視圖,多視圖學(xué)習(xí)也是機器學(xué)習(xí)過程中的重點研究課題。若是能在多視圖數(shù)據(jù)之間發(fā)現(xiàn)他們隱藏的互補性關(guān)系,那么就可以在很大程度上提高學(xué)習(xí)的效果。然而隨著現(xiàn)代社會技術(shù)的發(fā)展,數(shù)據(jù)的大規(guī)模應(yīng)用加大了提取數(shù)據(jù)并進行標(biāo)記的難度。那么如何在這種環(huán)境下獲得數(shù)據(jù)之間多視圖關(guān)系,并以此選擇出最大相關(guān)和最小冗余的子集,這是本文研究的主要內(nèi)容。參照當(dāng)前計算及算法研究領(lǐng)域的最新進展,分析并研究出了一種基于多視圖的半監(jiān)督特征選擇算法方式。此算法不但能夠有效的提取多視圖之間的互補信息,而且可以分析不同視圖中各個特征間的冗余關(guān)系。結(jié)合少量標(biāo)記的數(shù)據(jù)信息和沒有標(biāo)記的數(shù)據(jù)信息,同時進行特征選擇和聚類學(xué)習(xí),從而解決部分標(biāo)記的多視圖數(shù)據(jù)。本文的研究工作主要有以下貢獻:(1)將本文構(gòu)建一種改進的并行SVM,基于w-model,采取多個SVM分類器并行計算數(shù)據(jù),此方法既確保分類器推廣性能又縮短訓(xùn)練時間。(2)在進行多視圖的特征選擇時,對每個視圖中各個特征間的冗余關(guān)系進行了綜合考慮。
[Abstract]:When dealing with computer problems, the data encountered in reality are generally high-dimensional, and there are a lot of irrelevant lengthy features. This brings some difficulties to the solution of practical problems. People have developed a feature selection algorithm. This method can effectively reduce the dimension of the data, and can directly select the optimal feature subset from the original features of the data. The research on this subject has become a hot research topic in the field of machine learning and data mining. Multi-view learning is also an important research topic in the process of machine learning. Then we can improve the effect of learning to a great extent. However, with the development of modern society and technology, The large-scale application of data makes it more difficult to extract and mark data. In this environment, how to obtain the multi-view relationship between the data and select the subset of maximum correlation and minimal redundancy, This is the main content of this paper. This paper analyzes and studies a semi-supervised feature selection algorithm based on multi-view, which not only can extract the complementary information between multi-views, In addition, we can analyze the redundant relations among different features in different views, combine a small amount of tagged data information and unmarked data information, and carry out feature selection and clustering learning at the same time. The main contributions of this paper are as follows: 1) this paper constructs an improved parallel SVM, which is based on w-model, and uses several SVM classifiers to compute the data in parallel. This method not only ensures the generalization performance of classifier but also shortens the training time. 2) in the feature selection of multiple views, the redundant relations among each feature in each view are considered synthetically.
【學(xué)位授予單位】：山東師范大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP181;TP391.1

【參考文獻】

相關(guān)期刊論文 前10條

1 張德軍;何發(fā)智;袁志勇;石強;;基于模糊粗糙集的半監(jiān)督影像分類算法[J];華中科技大學(xué)學(xué)報(自然科學(xué)版);2016年01期

2 謝乃俊;楊國亮;羅璐;梁禮明;;基于低秩評分的非監(jiān)督特征選擇算法[J];計算機工程與設(shè)計;2015年06期

3 代琨;于宏毅;仇文博;李青;;基于SVM的網(wǎng)絡(luò)數(shù)據(jù)無監(jiān)督特征選擇算法[J];吉林大學(xué)學(xué)報(工學(xué)版);2015年02期

4 汪荊琪;徐林莉;;一種基于多視圖數(shù)據(jù)的半監(jiān)督特征選擇和聚類算法[J];數(shù)據(jù)采集與處理;2015年01期

5 林榮強;李鷗;李青;李林林;;基于類標(biāo)記擴展的半監(jiān)督網(wǎng)絡(luò)流量特征選擇算法[J];計算機應(yīng)用;2014年11期

6 王亦兵;潘志松;吳君青;賈波;胡谷雨;;基于鏈接關(guān)系的半監(jiān)督特征選擇算法[J];模式識別與人工智能;2014年02期

7 A.Y.Galashev;;Computer study of the spectral characteristics of the disperse water methane system[J];Chinese Physics B;2013年12期

8 Alexander Galashev;;Computer study of the formation of water-ammonia clusters and their dielectric properties[J];Chinese Physics B;2013年07期

9 李平紅;王勇;陶曉玲;;基于成對約束擴展的半監(jiān)督網(wǎng)絡(luò)流量特征選擇算法[J];傳感器與微系統(tǒng);2013年05期

10 王博;黃九鳴;賈焰;楊樹強;;適用于多種監(jiān)督模型的特征選擇方法研究[J];計算機研究與發(fā)展;2010年09期

，

本文編號：1641539