【摘要】：在模式分類問題中,數(shù)據(jù)往往存在許多不相關(guān)或是冗余的特征,從而影響分類的準(zhǔn)確性。特征選擇作為解決這一問題的有效手段,一直以來都是機(jī)器學(xué)習(xí)中的熱點(diǎn)。隨著數(shù)據(jù)規(guī)模的增加,原始的特征選擇方法已經(jīng)不滿足要求。特征選擇可以視為一個(gè)動(dòng)態(tài)尋優(yōu)的過程,而粒子群優(yōu)化算法是目前群體智能算法中的一個(gè)熱門的算法,由于其簡(jiǎn)單、易實(shí)現(xiàn)、尋優(yōu)效率高等特點(diǎn)受到了廣泛的關(guān)注。粒子群優(yōu)化算法與特征選擇方法的結(jié)合也成為了一個(gè)研究熱點(diǎn)。大量的研究表明了基于粒子群優(yōu)化算法與特征選擇結(jié)合是可行的,并且有著良好的性能表現(xiàn)。本文主要在粒子群優(yōu)化算法本身的改進(jìn)和特征選擇問題與粒子群優(yōu)化方法的結(jié)合兩個(gè)方面做了一定的工作。首先是對(duì)粒子群算法的改進(jìn),普通的粒子群算法由于其局限性,往往容易陷入局部最優(yōu),在骨干粒子群算法的基礎(chǔ)上,提出一種基于干擾因子的自適應(yīng)粒子群算法,在算法的初始過程中引入混沌模型增加初始粒子的多樣性,同時(shí)在更新機(jī)制中引入自適應(yīng)因子增加其全局搜索能力,提高算法的尋優(yōu)效率。其次改進(jìn)粒子群算法中粒子的局部和全局最優(yōu)的迭代公式。在更新過程中引入對(duì)于特征數(shù)目的討論,特別是在解碼過程中引入互信息篩選特征,精簡(jiǎn)特征子集。特征選擇的目的在于利用最少的特征達(dá)到最佳的優(yōu)化效果,以往的研究過程中,因?yàn)樽非蟾玫姆诸愋Ч?而忽視特征子集中的特征數(shù)目。最后提出一種基于混合模式評(píng)價(jià)機(jī)制的特征選擇算法。將特征選擇過程分為兩個(gè)階段,第一階段采用基于粗糙集的過濾模式評(píng)價(jià)機(jī)制,第二階段采用基于鄰近算法的封裝模式評(píng)價(jià)機(jī)制。為了驗(yàn)證上述提出的理論,選擇不同類型的數(shù)據(jù)集上進(jìn)行分類實(shí)驗(yàn),得到的實(shí)驗(yàn)結(jié)果驗(yàn)證了所提出算法的有效性和實(shí)用性。
[Abstract]:In the problem of pattern classification, there are often many unrelated or redundant features in the data, which affects the accuracy of classification. As an effective means to solve this problem, feature selection has always been a hot spot in machine learning. With the increase of data scale, the original feature selection method no longer meets the requirements. Feature selection can be regarded as a dynamic optimization process, and particle swarm optimization algorithm is a hot algorithm in swarm intelligence algorithm at present. because of its simplicity, easy implementation and high efficiency, particle swarm optimization algorithm has attracted extensive attention. The combination of particle swarm optimization algorithm and feature selection method has also become a research focus. A large number of studies have shown that the combination of particle swarm optimization algorithm and feature selection is feasible and has good performance. In this paper, some work has been done on the improvement of particle swarm optimization algorithm itself and the combination of feature selection problem and particle swarm optimization method. The first is to improve the particle swarm optimization algorithm. Because of its limitations, the ordinary particle swarm optimization algorithm is often easy to fall into local optimization. On the basis of the backbone particle swarm optimization algorithm, an adaptive particle swarm optimization algorithm based on interference factor is proposed. In the initial process of the algorithm, chaos model is introduced to increase the diversity of the initial particles, and at the same time, the adaptive factor is introduced into the update mechanism to increase its global search ability. Improve the optimization efficiency of the algorithm. Secondly, the local and global optimal iterative formulas of particles in particle swarm optimization are improved. In the process of updating, the discussion of the number of features is introduced, especially the mutual information filtering features are introduced in the decoding process to simplify the feature subset. The purpose of feature selection is to achieve the best optimization effect by using the least features. In the previous research process, the number of features in the feature subset was ignored because of the pursuit of better classification effect. Finally, a feature selection algorithm based on hybrid pattern evaluation mechanism is proposed. The feature selection process is divided into two stages. In the first stage, the filtering mode evaluation mechanism based on rough set is adopted, and in the second stage, the encapsulation mode evaluation mechanism based on neighborhood algorithm is adopted. In order to verify the above theory, different types of data sets are selected for classification experiments, and the experimental results verify the effectiveness and practicability of the proposed algorithm.
【學(xué)位授予單位】：南京郵電大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP18

【參考文獻(xiàn)】

相關(guān)期刊論文 前9條

1 鄢夢(mèng)迪;秦琳琳;吳剛;;基于主成分分析和K近鄰的文件類型識(shí)別算法[J];計(jì)算機(jī)應(yīng)用;2016年11期

2 王曉梅;林曉惠;黃鑫;;基于特征有效范圍的前向特征選擇及融合分類算法[J];小型微型計(jì)算機(jī)系統(tǒng);2016年06期

3 楊云峰;;基于Relief-SBS特征選擇算法的入侵檢測(cè)方法研究[J];河池學(xué)院學(xué)報(bào);2013年02期

4 姚旭;王曉丹;張玉璽;權(quán)文;;特征選擇方法綜述[J];控制與決策;2012年02期

5 麥范金;李東普;岳曉光;;基于雙向匹配法和特征選擇算法的中文分詞技術(shù)研究[J];昆明理工大學(xué)學(xué)報(bào)(自然科學(xué)版);2011年01期

6 葉吉祥;龔希齡;;一種快速的Wrapper式特征子集選擇新方法[J];長(zhǎng)沙理工大學(xué)學(xué)報(bào)(自然科學(xué)版);2010年04期

7 田東平;趙天緒;;基于Sigmoid慣性權(quán)值的自適應(yīng)粒子群優(yōu)化算法[J];計(jì)算機(jī)應(yīng)用;2008年12期

8 張浩;沈繼紅;張鐵男;李陽(yáng);;一種基于混沌映射的粒子群優(yōu)化算法及性能仿真[J];系統(tǒng)仿真學(xué)報(bào);2008年20期

9 張麗新;王家欽;趙雁南;楊澤紅;;機(jī)器學(xué)習(xí)中的特征選擇[J];計(jì)算機(jī)科學(xué);2004年11期

相關(guān)碩士學(xué)位論文 前6條

1 劉星;基于粒子群優(yōu)化算法的特征選擇方法研究[D];南京大學(xué);2015年

2 許露;基于SVM-RFE和粒子群算法的特征選擇算法研究[D];湖南師范大學(xué);2014年

3 張同偉;基于多分類器組合的垃圾網(wǎng)頁(yè)的檢測(cè)[D];華南理工大學(xué);2010年

4 姜百寧;機(jī)器學(xué)習(xí)中的特征選擇算法研究[D];中國(guó)海洋大學(xué);2009年

5 黃繼紅;基于改進(jìn)PSO的BP網(wǎng)絡(luò)的研究及應(yīng)用[D];長(zhǎng)沙理工大學(xué);2008年

6 付延安;基于遺傳算法與粗集理論的車間調(diào)度研究[D];山東大學(xué);2007年

，

本文編號(hào)：2504507