本文選題：FCM算法　切入點(diǎn)：粒子群算法　出處：《西南大學(xué)》2017年碩士論文

【摘要】：聚類分析的目標(biāo)是在相似的基礎(chǔ)上收集數(shù)據(jù)進(jìn)行分類,使得各個(gè)類之間的數(shù)據(jù)差別應(yīng)盡可能大,類內(nèi)之間的數(shù)據(jù)差別應(yīng)盡可能小,即為算法的選擇取決于數(shù)據(jù)的類型、聚類的目的和應(yīng)用方向。例如k-means、BIRCH、CURE、DBSCAN、COBWEB等,對(duì)于相同的數(shù)據(jù)集,使用不同的聚類算法可能有不同的劃分結(jié)果。FCM算法是目前應(yīng)用最為廣泛的聚類算法。研究發(fā)現(xiàn),傳統(tǒng)FCM算法存在兩個(gè)不足:第一,算法從樣本點(diǎn)出發(fā),通過(guò)優(yōu)化目標(biāo)函數(shù)計(jì)算各樣本點(diǎn)對(duì)于類中心的隸屬度,從而達(dá)到自動(dòng)分類的目的,如果初始值選擇不當(dāng)就會(huì)導(dǎo)致算法收斂到局部極小點(diǎn);第二,聚類分析處理數(shù)據(jù)樣本的各維屬性貢獻(xiàn)度是不一樣的,FCM算法采用標(biāo)準(zhǔn)的歐式距離進(jìn)行計(jì)算忽略了屬性權(quán)重值對(duì)聚類結(jié)果的影響。因此從本質(zhì)上來(lái)講,FCM算法是一種局部搜索的優(yōu)化算法�；谝陨戏治�,論文提出了基于屬性權(quán)重的混合聚類算法。主要研究?jī)?nèi)容如下:(1)將“粒子演化”策略結(jié)合“粒子分組及重組”引入粒子群算法,得到改進(jìn)的粒子群優(yōu)化算法,為屬性權(quán)重的求取提供了算法基礎(chǔ)。(2)屬性權(quán)重學(xué)習(xí)算法實(shí)現(xiàn):在改進(jìn)的粒子群優(yōu)化算法中,將粒子的位置向量作為屬性權(quán)重向量,選用交叉熵作為屬性權(quán)重評(píng)價(jià)函數(shù),利用梯度下降法極小化屬性權(quán)重評(píng)價(jià)函數(shù),通過(guò)迭代最終得到一組最優(yōu)的屬性權(quán)重值。(3)混合聚類算法實(shí)現(xiàn):將遺傳算法和模擬退火算法相結(jié)合,引入FCM聚類算法,初始聚類中心映射成染色體,目標(biāo)函數(shù)作為遺傳算法的適應(yīng)度函數(shù),經(jīng)過(guò)選擇、交叉和變異,運(yùn)用FCM聚類算法計(jì)算聚類中心、隸屬度及個(gè)體適應(yīng)度值,利用模擬退火算法以一定概率接受新個(gè)體,通過(guò)迭代,最終得到全局最優(yōu)解。
[Abstract]:The goal of clustering analysis is to collect data for classification on a similar basis, so that the data differences between each class should be as large as possible, and the data differences between classes should be as small as possible, that is, the choice of algorithm depends on the type of data. For the same data set, different clustering algorithms may have different partition results. FCM algorithm is the most widely used clustering algorithm. The traditional FCM algorithm has two disadvantages: first, the algorithm calculates the membership degree of each sample point to the center of the class by optimizing the objective function from the sample point, so as to achieve the purpose of automatic classification. If the initial value is not chosen properly, the algorithm will converge to a local minimum. Second, In clustering analysis, the contribution of attributes in different dimensions of data samples is different. The standard Euclidean distance algorithm is used to calculate and ignore the influence of attribute weights on clustering results. Therefore, FCM algorithm is essentially a kind of FCM algorithm. Local search optimization algorithm. Based on the above analysis, In this paper, a hybrid clustering algorithm based on attribute weight is proposed. The main research contents are as follows: (1) the particle evolution strategy combined with "particle grouping and recombination" is introduced into the particle swarm optimization algorithm, and the improved particle swarm optimization algorithm is obtained. In the improved particle swarm optimization algorithm, the particle position vector is used as the attribute weight vector, and the cross-entropy is chosen as the attribute weight evaluation function. Using gradient descent method to minimize the attribute weight evaluation function, a group of optimal attribute weight value. 3) hybrid clustering algorithm is obtained by iteration. The genetic algorithm is combined with simulated annealing algorithm, and the FCM clustering algorithm is introduced. The initial cluster center is mapped to chromosome, and the objective function is used as the fitness function of genetic algorithm. After selection, crossover and mutation, FCM clustering algorithm is used to calculate the clustering center, membership degree and individual fitness. The simulated annealing algorithm is used to accept the new individuals with a certain probability, and the global optimal solution is obtained by iteration.
【學(xué)位授予單位】：西南大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP311.13

【參考文獻(xiàn)】

相關(guān)期刊論文 前8條

1 趙乃剛;鄧景順;;粒子群優(yōu)化算法綜述[J];科技創(chuàng)新導(dǎo)報(bào);2015年26期

2 解晨;韋雄奕;;模擬退火算法和遺傳算法的比較與思考[J];電腦知識(shí)與技術(shù);2013年19期

3 任妮;金剛;;基于模擬退火算法的PID參數(shù)優(yōu)化設(shè)計(jì)[J];云南電力技術(shù);2013年02期

4 徐章艷;侯偉;宋威;楊炳儒;;一個(gè)有效的基于信息熵的啟發(fā)式屬性約簡(jiǎn)算法[J];小型微型計(jì)算機(jī)系統(tǒng);2009年09期

5 楊春宇;周杰;;一種混合屬性數(shù)據(jù)流聚類算法[J];計(jì)算機(jī)學(xué)報(bào);2007年08期

6 王麗娟;關(guān)守義;王曉龍;王熙照;;基于屬性權(quán)重的Fuzzy C Mean算法[J];計(jì)算機(jī)學(xué)報(bào);2006年10期

7 劉英;;遺傳算法與改進(jìn)的FCM聚類算法的結(jié)合[J];甘肅科技;2006年03期

8 王熙照,王麗娟,王利偉;傳遞閉包聚類中的模糊性分析[J];計(jì)算機(jī)工程與應(yīng)用;2003年18期

相關(guān)碩士學(xué)位論文 前2條

1 陳漠;用于最優(yōu)化問(wèn)題的改進(jìn)粒子群優(yōu)化算法研究[D];吉林大學(xué);2015年

2 張必蘭;改進(jìn)的粒子群優(yōu)化算法及應(yīng)用研究[D];重慶大學(xué);2007年

，

本文編號(hào)：1659253