本文選題：關(guān)聯(lián)規(guī)則 + 商品銷售　； 參考：《湖北大學(xué)》2014年碩士論文

【摘要】：隨著數(shù)據(jù)挖掘技術(shù)不斷發(fā)展,數(shù)據(jù)挖掘帶來的商業(yè)價(jià)值越來越受到各行業(yè)的關(guān)注,急切想借助數(shù)據(jù)挖掘東風(fēng)創(chuàng)造更多價(jià)值。衡量一個(gè)企業(yè)創(chuàng)造更多價(jià)值的直接標(biāo)準(zhǔn)是其商品銷售額和銷售量的增加。關(guān)聯(lián)規(guī)則是數(shù)據(jù)挖掘應(yīng)用最廣泛的算法之一,關(guān)聯(lián)規(guī)則在商品銷售中的應(yīng)用雖然很早就有成功案例,但隨著商業(yè)競爭的日益激烈,如何提高商品推薦的有效性成了亟待解決的難題。 圍繞提高商品銷售的有效性問題,本文主要從以下幾個(gè)方面開展工作： 首先,在廣泛查閱有關(guān)資料的基礎(chǔ)上,深入研究了已有的關(guān)聯(lián)規(guī)則挖掘技術(shù)和聚類方法。在此基礎(chǔ)上,通過大量實(shí)驗(yàn),選取出適合商品銷售推薦的關(guān)聯(lián)規(guī)則挖掘算法和聚類算法； 其次,利用選出的聚類算法,提取用戶特征,對(duì)客戶進(jìn)行分類；在分類客戶數(shù)據(jù)的基礎(chǔ)上,將商品銷售數(shù)據(jù)集進(jìn)行轉(zhuǎn)換,形成矢量化的交易數(shù)據(jù)庫,便于后期挖掘效率的提高； 然后,利用選出的關(guān)聯(lián)規(guī)則挖掘算法對(duì)轉(zhuǎn)換數(shù)據(jù)庫進(jìn)行挖掘,得出針對(duì)不同類型客戶的挖掘結(jié)果�；谏鲜鏊悸�,本文提出了一個(gè)CAM(Cluster-Association Mining)算法； 最后,將提出的算法在真實(shí)數(shù)據(jù)集上予以了實(shí)現(xiàn)。結(jié)果表明,本文提出的算具有較強(qiáng)的針對(duì)性和有效性。
[Abstract]:With the development of data mining technology, the commercial value brought by data mining is paid more and more attention by various industries, and it is eager to create more value with the help of data mining. The direct measure of an enterprise's creation of more value is the increase in its merchandise sales and sales. Association rules is one of the most widely used algorithms in data mining. Although the application of association rules in commodity sales has been successful for a long time, but with the increasingly fierce business competition, How to improve the effectiveness of commodity recommendation has become a difficult problem to be solved. Focusing on improving the effectiveness of commodity sales, this paper mainly from the following aspects of work: Firstly, the existing association rule mining techniques and clustering methods are studied on the basis of extensive reference to relevant data. On this basis, through a large number of experiments, the association rules mining algorithm and clustering algorithm suitable for commodity sales recommendation are selected. Secondly, the selected clustering algorithm is used to extract the user features and classify the customers. On the basis of the classified customer data, the commodity sales data set is transformed to form a vectorized transaction database, which is convenient to improve the efficiency of mining in the later stage. Then, mining the conversion database with the selected association rules mining algorithm, and obtaining the mining results for different types of customers. Based on the above ideas, a CAM(Cluster-Association mining algorithm is proposed in this paper. Finally, the proposed algorithm is implemented on the real data set. The results show that the proposed algorithm has strong pertinence and effectiveness.
【學(xué)位授予單位】：湖北大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類號(hào)】：F721;F224

【參考文獻(xiàn)】

相關(guān)期刊論文 前6條

1 馬占欣;王新社;黃維通;陸玉昌;;對(duì)最小置信度門限的置疑[J];計(jì)算機(jī)科學(xué);2007年06期

2 鄒筱梅,姜山,唐賢瑛;基于決策樹的股市數(shù)據(jù)挖掘與仿真[J];計(jì)算機(jī)仿真;2004年03期

3 孫吉貴;劉杰;趙連宇;;聚類算法研究[J];軟件學(xué)報(bào);2008年01期

4 雷松澤,郝艷;基于決策樹的就業(yè)數(shù)據(jù)挖掘[J];西安工業(yè)學(xué)院學(xué)報(bào);2005年05期

5 彭松波;何文秀;;決策樹在高校就業(yè)管理系統(tǒng)中的應(yīng)用研究[J];中原工學(xué)院學(xué)報(bào);2006年04期

6 晏杰;亓文娟;;基于Aprior&FP-growth算法的研究[J];計(jì)算機(jī)系統(tǒng)應(yīng)用;2013年05期

相關(guān)博士學(xué)位論文 前1條

1 劉智;關(guān)聯(lián)規(guī)則挖掘方法及其在冠心病中醫(yī)診療中的應(yīng)用研究[D];大連海事大學(xué);2012年

，

本文編號(hào)：1946588