【摘要】：聚類算法是非監(jiān)督學(xué)習(xí)算法領(lǐng)域中的一個(gè)經(jīng)典算法,在對(duì)數(shù)據(jù)集的分布情況沒(méi)有先驗(yàn)知識(shí)了解的前提下,利用相似性度量將數(shù)據(jù)集進(jìn)行分類。通過(guò)聚類,發(fā)現(xiàn)數(shù)據(jù)樣本間的關(guān)聯(lián)關(guān)系。然而,在算法性能方面,傳統(tǒng)聚類算法的表現(xiàn)相對(duì)較差。因此,利用群智能優(yōu)化的不同參數(shù)和啟發(fā)式函數(shù)規(guī)則來(lái)提高聚類效果和魯棒性越來(lái)越受到研究者的關(guān)注。近年新興的組群搜索優(yōu)化算法,具有參數(shù)少,操作簡(jiǎn)單,不易陷入局部最優(yōu)等優(yōu)勢(shì),同時(shí)也面臨收斂速度較慢等問(wèn)題。本文提出改進(jìn)組群搜索優(yōu)化算法,將其思想應(yīng)用于求解聚類算法,提出四種基于組群搜索優(yōu)化思想的聚類算法,將傳統(tǒng)聚類問(wèn)題作為優(yōu)化問(wèn)題來(lái)解決。主要?jiǎng)?chuàng)新點(diǎn)如下:第一,提出基于發(fā)現(xiàn)者適應(yīng)度范圍排序策略的DRGSO算法。引入排序策略,避免陷入局部最優(yōu),克服了傳統(tǒng)組群搜索優(yōu)化算法(Group Search Optimizer Algorithm,GSO)中組群算子搜索方向單一,資源信息利用不足的問(wèn)題,豐富尋優(yōu)啟發(fā)式搜索的資源信息,提高發(fā)現(xiàn)者種群的全局搜索能力。在11組國(guó)際標(biāo)準(zhǔn)測(cè)試函數(shù)上的仿真實(shí)驗(yàn)結(jié)果表明:DRGSO算法性能明顯提高,收斂速度較快。第二,提出基于差分演化策略的組群搜索優(yōu)化算法(Differential Ranking-based Group Search Optimizer,DRGSO)。引入四種差分變異算子模式來(lái)設(shè)計(jì)組群算子,加快算法收斂速度,克服了 GSO算法存在的收斂速度較慢,精度不高等問(wèn)題。與GSO相比,改進(jìn)的組群搜索操作使發(fā)現(xiàn)者搜索結(jié)構(gòu)更為優(yōu)化,能確保算法搜索路徑多樣性,提高算法收斂精度。第三,提出四種基于組群優(yōu)化的聚類分析方法,具體包括:GSO聚類算法、基于均值GSO聚類算法、DRGSO聚類算法和基于均值DRGSO聚類算法。首先,利用GSO優(yōu)化算法的種群結(jié)構(gòu)來(lái)編碼簇中心位置,優(yōu)化聚類算法的簇分配過(guò)程,在GSO迭代過(guò)程中完成簇移動(dòng),簡(jiǎn)化聚類分析方法的復(fù)雜度,提高聚類效果。其次,提出根據(jù)發(fā)現(xiàn)者種群適應(yīng)度范圍進(jìn)行局部均值策略,避免陷入局部最優(yōu),完善個(gè)體間信息資源的共享模式。最后,利用差分變異算子模式,讓簇移動(dòng)更具多樣性,提高聚類算法的全局搜索能力。在國(guó)際標(biāo)準(zhǔn)數(shù)據(jù)集Iris、Wine上的實(shí)驗(yàn)結(jié)果表明,組群搜索優(yōu)化聚類算法的聚類效果更明顯、具有較好穩(wěn)定性和魯棒性。
[Abstract]:Clustering algorithm is a classical algorithm in the field of unsupervised learning algorithm. In the absence of prior knowledge of the distribution of data sets, the data sets are classified by similarity measurement. By clustering, the correlation between data samples is found. However, the performance of traditional clustering algorithm is relatively poor. Therefore, using different parameters and heuristic function rules of swarm intelligence optimization to improve clustering effect and robustness has attracted more and more attention. In recent years, the new cluster search optimization algorithm has the advantages of few parameters, simple operation, not easy to fall into local optimum, and also faces some problems such as slow convergence speed and so on. In this paper, an improved cluster search optimization algorithm is proposed, and its idea is applied to the clustering algorithm. Four clustering algorithms based on cluster search optimization are proposed. The traditional clustering problem is solved as an optimization problem. The main innovations are as follows: first, the DRGSO algorithm based on the range ordering strategy of discoverer fitness is proposed. The sorting strategy is introduced to avoid falling into local optimum, which overcomes the problem of single search direction of group operator and insufficient utilization of resource information in traditional group search optimization algorithm (Group Search Optimizer Algorithm,GSO), and enriches the resource information of heuristic search. Improve the global search ability of the discoverer population. The simulation results on 11 international standard test functions show that the performance of DRGSO algorithm is obviously improved and the convergence rate is faster. Secondly, a group search optimization algorithm (Differential Ranking-based Group Search Optimizer,DRGSO) based on differential evolution strategy is proposed. Four kinds of differential mutation operator patterns are introduced to design group operators to accelerate the convergence speed of the algorithm and overcome the problems of slow convergence rate and low precision of GSO algorithm. Compared with GSO, the improved group search operation can optimize the search structure of the discoverer, ensure the diversity of the search path and improve the convergence accuracy of the algorithm. Thirdly, four clustering methods based on cluster optimization are proposed, including: GSO clustering algorithm, GSO clustering algorithm based on mean value, DRGSO clustering algorithm and DRGSO clustering algorithm based on mean value. Firstly, the cluster center position is encoded by the population structure of GSO optimization algorithm, and the cluster assignment process is optimized. The cluster movement is completed in the GSO iteration process, which simplifies the complexity of the clustering analysis method and improves the clustering effect. Secondly, according to the range of population fitness of discoverers, a local mean strategy is proposed to avoid falling into local optimum and perfect the sharing mode of information resources among individuals. Finally, the differential mutation operator pattern is used to make the cluster move more diverse and improve the global searching ability of the clustering algorithm. The experimental results on the international standard data set Iris,Wine show that the clustering effect of cluster search optimization clustering algorithm is more obvious, and the clustering algorithm has better stability and robustness.
【學(xué)位授予單位】：天津科技大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP18;TP311.13

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 金建國(guó);;聚類方法綜述[J];計(jì)算機(jī)科學(xué);2014年S2期

2 汪慎文;丁立新;張文生;郭肇祿;謝承旺;;差分進(jìn)化算法研究進(jìn)展[J];武漢大學(xué)學(xué)報(bào)(理學(xué)版);2014年04期

3 肖立中;劉云翔;陳麗瓊;;基于改進(jìn)粒子群的加速K均值算法在入侵檢測(cè)中的研究[J];系統(tǒng)仿真學(xué)報(bào);2014年08期

4 劉華臣;王錫淮;肖健梅;王海鋒;;基于群搜索算法的電力系統(tǒng)無(wú)功優(yōu)化[J];電力系統(tǒng)保護(hù)與控制;2014年14期

5 趙新超;劉子陽(yáng);;差分和擾動(dòng)混合的多策略粒子群優(yōu)化算法[J];計(jì)算機(jī)科學(xué)與探索;2014年02期

6 于海濤;賈美娟;王慧強(qiáng);邵國(guó)強(qiáng);;基于人工魚(yú)群的優(yōu)化K-means聚類算法[J];計(jì)算機(jī)科學(xué);2012年12期

7 汪慎文;丁立新;謝承旺;謝大同;舒萬(wàn)能;楊華;;群搜索優(yōu)化算法中角色分配策略的研究[J];小型微型計(jì)算機(jī)系統(tǒng);2012年09期

8 汪慎文;丁立新;謝大同;舒萬(wàn)能;謝承旺;楊華;;應(yīng)用反向?qū)W習(xí)策略的群搜索優(yōu)化算法[J];計(jì)算機(jī)科學(xué);2012年09期

9 王千;王成;馮振元;葉金鳳;;K-means聚類算法研究綜述[J];電子設(shè)計(jì)工程;2012年07期

10 陳小全;張繼紅;;基于改進(jìn)粒子群算法的聚類算法[J];計(jì)算機(jī)研究與發(fā)展;2012年S1期

相關(guān)博士學(xué)位論文 前3條

1 劉逸;粒子群優(yōu)化算法的改進(jìn)及應(yīng)用研究[D];西安電子科技大學(xué);2013年

2 曲建華;基于群體智能的聚類分析[D];山東師范大學(xué);2010年

3 陳毅恒;文本檢索結(jié)果聚類及類別標(biāo)簽抽取技術(shù)研究[D];哈爾濱工業(yè)大學(xué);2010年

相關(guān)碩士學(xué)位論文 前7條

1 劉嬋;蟻群與K均值聚類算法融合研究及其在用戶分群中的應(yīng)用[D];西南科技大學(xué);2015年

2 鄭玉艷;基于粒子群優(yōu)化的聚類算法研究[D];山東師范大學(xué);2015年

3 周虹;基于自適應(yīng)粒子群的k-中心聚類算法研究[D];長(zhǎng)沙理工大學(xué);2012年

4 房娟艷;混合群搜索優(yōu)化算法及其應(yīng)用研究[D];太原科技大學(xué);2010年

5 劉壽吉;基于群體智能的蟻群聚類算法及應(yīng)用[D];復(fù)旦大學(xué);2008年

6 馬學(xué)陽(yáng);基于粒子群聚類方法的上市公司財(cái)務(wù)能力評(píng)價(jià)模型[D];吉林大學(xué);2007年

7 陳君彥;基于粒子群的聚類算法改進(jìn)及其在訪問(wèn)模式中的應(yīng)用研究[D];天津大學(xué);2007年

，

本文編號(hào)：2391566