本文關(guān)鍵詞： 數(shù)據(jù)挖掘 聚類分析 數(shù)據(jù)化運(yùn)營(yíng) 營(yíng)銷策略　出處：《華南理工大學(xué)》2015年碩士論文　論文類型：學(xué)位論文

【摘要】：聚類分析是數(shù)據(jù)挖掘的重要功能之一,近年來(lái)在該領(lǐng)域的研究取得了長(zhǎng)足的發(fā)展。聚類分析方法所涉及的領(lǐng)域幾乎遍及人工智能的方方面面,在各行各業(yè)以信息分析為基礎(chǔ)的決策支持系統(tǒng)活動(dòng)中扮演越來(lái)越重要的角色。它在電子商務(wù)、圖像處理,模式識(shí)別、文本分類等領(lǐng)域有廣泛的應(yīng)用。本文對(duì)聚類分析算法在客戶分類中的應(yīng)用進(jìn)行了深入的研究,主要研究工作如下:1)深入研究聚類分析技術(shù),對(duì)聚類分析中的各種算法進(jìn)行了詳細(xì)的分析。2)重點(diǎn)研究聚類分析算法中應(yīng)用比較廣泛的系統(tǒng)聚類算法。對(duì)過(guò)比較系統(tǒng)聚類算法與其它聚類算法的優(yōu)缺點(diǎn),分析各個(gè)方法的適用性,以及系統(tǒng)聚類在客戶分類中的適用性。3)在深入研究了系統(tǒng)聚類算法的基礎(chǔ)上,對(duì)H銀行網(wǎng)上轉(zhuǎn)賬系統(tǒng)的客戶進(jìn)行分類。并對(duì)分類過(guò)程中所涉及的指標(biāo)參數(shù)變量選取、距離的度量、數(shù)據(jù)市集的建立以及數(shù)據(jù)預(yù)處理等方面進(jìn)行了詳細(xì)的論述。4)對(duì)使用系統(tǒng)聚類算法進(jìn)行客戶分類后的結(jié)果敘述驗(yàn)證,最后結(jié)果實(shí)踐證明該聚類算法在客戶分類上的有效性。通過(guò)對(duì)分類后篩選出來(lái)的有價(jià)值客戶群體進(jìn)行定向營(yíng)銷,付費(fèi)客戶數(shù)量的提升率比起在測(cè)試運(yùn)營(yíng)階段有所提升。本文將數(shù)據(jù)挖掘中的聚類分析引入到銀行的營(yíng)銷策略分析中來(lái),為市場(chǎng)的營(yíng)銷戰(zhàn)略與策略提供了非�？茖W(xué)的參考體系,也為各類企業(yè)在數(shù)據(jù)化運(yùn)營(yíng)的方向上提供了非常有價(jià)值的實(shí)踐。
[Abstract]:Clustering analysis is one of the important functions of data mining. In recent years, great progress has been made in the research in this field. Clustering analysis involves almost every aspect of artificial intelligence. It plays an increasingly important role in the activities of decision support systems based on information analysis in various industries. It plays an increasingly important role in electronic commerce, image processing, and pattern recognition. Text classification and other fields have a wide range of applications. In this paper, the application of clustering analysis algorithm in customer classification has been deeply studied, the main research work is as follows: 1) deeply study clustering analysis technology. This paper makes a detailed analysis of all kinds of algorithms in clustering analysis. (2) focusing on the system clustering algorithm which is widely used in clustering analysis algorithm, the advantages and disadvantages of over-comparison system clustering algorithm and other clustering algorithms are discussed. Analysis of the applicability of each method, and the applicability of system clustering in customer classification. 3) on the basis of in-depth study of the system clustering algorithm. The customer of H bank online transfer system is classified, and the parameter variables are selected and the distance is measured in the process of classification. The establishment of the data market and data preprocessing are discussed in detail. 4) the results of customer classification using the system clustering algorithm are described and verified. Finally, the practice proves the effectiveness of the clustering algorithm in customer classification. Through the classification of valuable customer groups selected out of the targeted marketing. The increase rate of the number of paying customers is higher than that in the test operation phase. This paper introduces the clustering analysis of data mining into the marketing strategy analysis of banks. It provides a very scientific reference system for marketing strategy and strategy, and also provides a very valuable practice for all kinds of enterprises in the direction of data operation.
【學(xué)位授予單位】：華南理工大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2015
【分類號(hào)】：TP311.13

【參考文獻(xiàn)】

相關(guān)期刊論文 前3條

1 肖娟;葉楓;;利用數(shù)據(jù)挖掘管理客戶關(guān)系[J];經(jīng)濟(jì)管理;2003年01期

2 丁劍敏;數(shù)據(jù)挖掘技術(shù)及其在商業(yè)銀行中的應(yīng)用[J];市場(chǎng)周刊(財(cái)經(jīng)論壇);2003年04期

3 張振華;數(shù)據(jù)倉(cāng)庫(kù)、數(shù)據(jù)挖掘在銀行中的應(yīng)用[J];中國(guó)金融電腦;1998年10期

，

本文編號(hào)：1448853