本文選題：物流服務　切入點：回購行為　出處：《浙江理工大學》2017年碩士論文　論文類型：學位論文

【摘要】：在網絡購物平臺中,已購顧客的物流服務評論通常影響著新顧客購買決策行為。作為評論內容的重要組成部分,物流服務評論如何影響顧客購買決策是電子商務企業(yè)十分關注的問題。根據中國電子商務投訴與維權公共服務平臺統(tǒng)計,發(fā)現(xiàn)物流快遞的投訴問題已經成為制約B2C企業(yè)發(fā)展的重要因素,如何提升物流服務質量,增加顧客滿意成為亟需解決的課題。為此,本文旨在通過抓取某電商女裝在線評論數(shù)據,提取出顧客最關注的物流服務要素以及常見的顧客購買后行為,找出兩者間的關聯(lián)關系,為B2C企業(yè)優(yōu)化物流服務提供新的路徑以及改進物流服務質量提供依據。首先,利用樂思信息采集系統(tǒng)抓取京東商城女裝的在線評論信息。接著清理采集到的數(shù)據,刪除缺失值,運用PROC正則表達式過濾廣告、視頻等干擾數(shù)據。然后,在ICTCLAS分詞系統(tǒng)的基礎上,添加用戶詞典,對采集到的數(shù)據進行分詞處理。根據得到的分詞結果,在現(xiàn)有的情感詞典本題庫基礎上,結合基于主題詞和情感詞的雙向迭代算法構建候選詞集,構建用戶情感詞典,利用微博情感計算系統(tǒng)判斷每條在線評論的情感極性,同時,得出詞頻統(tǒng)計結果。依據詞頻統(tǒng)計的結果,結合中國電子商務投訴與維權公共服務平臺以及物流服務質量相關文獻篩選出與物流服務質量相關詞匯的詞頻結果,同時篩選出與顧客購買后行為相關詞匯的詞頻結果,整理得出顧客最關注的物流服務要素和常見的顧客購買后行為。最后,依據分詞結果,提取有效的關鍵詞,結合物流服務要素和顧客購買后行為的劃分,建立在線評論關聯(lián)矩陣,運用FP-growth算法挖掘物流服務要素與顧客購買后行為間的強關聯(lián)規(guī)則。通過內容分析、關聯(lián)規(guī)則挖掘和分析,得到如下結論:(1)在線評論中顧客最關注的物流服務要素依次為:包裝完整性、發(fā)貨速度、運送速度、送貨速度、快遞員服務態(tài)度、售后服務態(tài)度和退換貨速度;(2)與顧客回購行為具有強關聯(lián)性物流服務要素有:發(fā)貨速度、運送速度、送貨速度、快遞員服務態(tài)度;(3)與顧客推薦行為具有強關聯(lián)性物流服務要素有:包裝完整、發(fā)貨速度、運送速度、送貨速度。研究結果有助于B2C企業(yè)尋找物流服務優(yōu)化的關鍵點,為改進物流服務質量提供借鑒。
[Abstract]:In the online shopping platform, the logistics service reviews of purchased customers usually affect the purchasing decision behavior of new customers. How logistics service reviews affect customer purchasing decisions is a matter of great concern to e-commerce enterprises. According to statistics of China's e-commerce complaints and rights protection public service platform, It is found that the complaint problem of logistics express has become an important factor restricting the development of B2C enterprises. How to improve the quality of logistics service and increase customer satisfaction has become an urgent problem. The purpose of this paper is to find out the relationship between the online review data of a certain e-commerce women's wear and the logistics service elements that customers are most concerned about and the common behavior of the customers after purchase, so as to find out the relationship between them. This paper provides a new way for B2C enterprises to optimize logistics service and improve the quality of logistics service. Firstly, the online comment information of women's wear of JingDong Mall is captured by using the information collection system of Le Si. Then the collected data is cleaned up, and the missing value is deleted. The PROC regular expression is used to filter the interfering data such as advertisement, video and so on. Then, on the basis of the ICTCLAS word segmentation system, the user dictionary is added to deal with the collected data. On the basis of the existing emotion dictionary, combined with the bidirectional iterative algorithm based on the subject word and emotion word, the candidate word set is constructed, the user emotion dictionary is constructed, and the emotion polarity of every online comment is judged by Weibo emotion calculation system. At the same time, According to the results of word frequency statistics, combined with the Chinese e-commerce complaints and rights protection public service platform, as well as logistics service quality related literature to screen out the word frequency results related to the quality of logistics services, At the same time, the frequency of words related to the behavior of customers is screened out, and the logistics service elements and common post-purchase behaviors of customers are sorted out. Finally, according to the segmentation results, the effective keywords are extracted. Combined with the division of logistics service elements and customers' post-purchase behavior, the online comment association matrix is established, and the strong association rules between logistics service elements and customers' post-purchase behavior are mined by using FP-growth algorithm, and the association rules are mined and analyzed by content analysis, association rules mining and analysis. The following conclusions are drawn as follows: (1) in the online review, the most important logistics service elements that customers are most concerned about are: package integrity, delivery speed, delivery speed, delivery speed, service attitude of couriers, The attitude of after-sales service and the speed of returning and exchanging goods have strong correlation with the behavior of customer repurchase. The factors of logistics service are: speed of delivery, speed of delivery, speed of delivery, speed of delivery, The service attitude of couriers has a strong correlation with customer recommendation. The key factors of logistics service are: package integrity, delivery speed, delivery speed, delivery speed. The research results are helpful for B2C enterprises to find the key points of logistics service optimization. To improve the quality of logistics services to provide reference.
【學位授予單位】：浙江理工大學
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：F724.6;F274;F713.55

【參考文獻】

相關期刊論文 前10條

1 李慧;劉貴全;瞿春燕;;頻繁和高效用項集挖掘[J];計算機科學;2015年05期

2 宗蕊;葛澤慧;;消費者對B2C網購物流服務因素的感知分析——基于京東商城在線客戶評論的實證研究[J];消費經濟;2014年02期

3 楊永清;于本海;;基于物流服務的消費者在線購買后行為實證分析[J];管理學報;2014年03期

4 崔廣風;鐘儀華;;數(shù)據挖掘方法在商業(yè)領域的應用研究綜述[J];現(xiàn)代商業(yè);2013年27期

5 王家琦;張耀荔;陳靜;;B2C網絡購物模式下物流服務質量與顧客滿意的關系研究[J];中國商貿;2013年01期

6 張成功;劉培玉;朱振方;方明;;一種基于極性詞典的情感分析方法[J];山東大學學報(理學版);2012年03期

7 嚴建援;張麗;張蕾;;電子商務中在線評論內容對評論有用性影響的實證研究[J];情報科學;2012年05期

8 趙鵬;趙志偉;卓景文;;一種情感詞語義加權的句子傾向性識別方法[J];計算機工程與應用;2011年35期

9 蔣園園;;B2C網絡購物物流服務質量評述[J];中國集體經濟;2011年16期

10 郎楊琴;孔麗華;;科學研究的第四范式 吉姆·格雷的報告“e-Science:一種科研模式的變革”簡介[J];科研信息化技術與應用;2010年02期

，

本文編號：1610809