本文選題：淘寶API + 屬性分類��； 參考：《內蒙古科技大學》2014年碩士論文

【摘要】：淘寶開放平臺是針對淘寶網(wǎng)各項商務業(yè)務開發(fā)出的開放平臺，它為開發(fā)者提供了參與淘寶各項應用的基礎材料，例如API、數(shù)據(jù)管理等。它的目標把淘寶網(wǎng)可以提供的許多服務及時方便的傳送給希望借助平臺對自己店鋪進行管理改進的商家和有能力借助平臺提供的服務進行研究開發(fā)的開發(fā)者。 淘寶網(wǎng)作為國內最大的電商網(wǎng)站，每月每天甚至每小時的交易量都是相當大的，而作為買家在淘寶網(wǎng)上購買商品之后，無論對商品滿意與否，都可以通過淘寶網(wǎng)提供的商品評價系統(tǒng)對所購買的商品做出自己的評價，這就產生了大量的評價信息。這些評價對今后購買同種商品的買家可以產生重要的指導作用；評價信息里反映出的買家對商品比較關注的一些問題也為賣家不斷改進商品提供了方向。 但是目前淘寶網(wǎng)的評價系統(tǒng)只是單純的將買家的評價羅列在產品信息頁面上，這些評價包含了買家對該商品在做工、款式、服務態(tài)度、發(fā)貨速度、物流速度等相關評價屬性的信息反饋，這對想查看該商品某種評價屬性信息的買家和賣家造成了很大的困難。有些商品評價信息較少，可以通過逐條查看的的方法來獲取需要的信息。但是對于有幾千條甚至上萬條評價信息的商品來說，逐條查看必定會耗費大量的精力，也無法取得較好的效果。同時，淘寶提供的好評、中評和差評都是由買家人工選擇的。有時候買家對商品并不很滿意，但是由于賣家服務態(tài)度較好等原因也給出了好評；而有些差評則是因為賣家服務態(tài)度不好或者物流送貨不及時等原因，，但是商品本身質量并沒有問題，這就對信息獲取造成了誤差。針對上面的問題設計并實現(xiàn)的淘寶商品評價屬性分類系統(tǒng)，利用淘寶開放平臺提供的API技術獲取相關評價信息，然后按質量、物流和服務這三項關注度最高的屬性將評價信息進行分類，并對每種屬性按評價內容情感傾向度分別判別出好評和差評。本系統(tǒng)實現(xiàn)了買家在淘寶網(wǎng)意向商品的大量評價中篩選出自己希望了解的關于商品某屬性的評價信息，同時也幫助賣家了解在售商品各個屬性的買家反饋信息，從而更好的改進商品。
[Abstract]:The open platform of Taobao is an open platform developed for various business business of Taobao network. It provides the basic materials for developers to participate in the various applications of Taobao, such as API, data management, etc. its goal is to deliver a lot of services provided by the Taobao network to hope to improve the management of its own shop with the help of the platform. Developers and developers who are capable of using platform services.
Taobao net, as the largest e-commerce website in China, is quite large every month or even per hour. As a buyer, after buying goods on the Taobao net, no matter whether the goods are satisfied or not, it can be evaluated by the commodity evaluation system provided by Taobao, which has produced a large number of comments. Price information. These evaluations have an important guiding role to buy buyers of the same commodity in the future; some of the problems that the buyers pay more attention to in the evaluation information also provide the direction for the seller to constantly improve the commodity.
However, at present, the evaluation system of Taobao network simply lists the buyer's evaluation on the product information page. These evaluations include the buyer's feedback on the evaluation attributes of the goods in workmanship, style, service attitude, delivery speed, logistics speed and so on. This is the buyer and seller of some evaluation attribute information for the goods. There is a great deal of difficulty. Some goods have less information and can get the information needed by the way to view each other. But for a commodity with thousands or even tens of thousands of evaluation information, it will cost a lot of energy and can not achieve better results. At the same time, Taobao offers good comments, comments and poor results. The evaluation is selected by the buyer. Sometimes the buyer is not very satisfied with the goods, but the seller's service attitude is good, and some bad reviews are because the seller's service attitude is not good or the logistics delivery is not timely and so on, but the quality of the goods itself is not a problem, which causes the information acquisition. Error. The Taobao commodity evaluation attribute classification system designed and implemented on the above problems, uses the API technology provided by Taobao open platform to obtain relevant evaluation information, then classifies the evaluation information according to the three attributes of the highest quality, logistics and service, and each attribute is judged by the emotional inclination of the evaluation content respectively. The system realizes the buyer to screen out the evaluation information about the property of the goods and help the seller to understand the feedback information of the buyers in the sale of the goods in a large amount of evaluation of the intentional goods of Taobao net, and to improve the goods better.
【學位授予單位】：內蒙古科技大學
【學位級別】：碩士
【學位授予年份】：2014
【分類號】：TP393.092

【參考文獻】

相關期刊論文 前6條

1 劉曉;莫倩;張政;;網(wǎng)絡評論觀點分類研究[J];北京工商大學學報(自然科學版);2008年03期

2 李實;葉強;李一軍;Rob Law;;中文網(wǎng)絡客戶評論的產品特征挖掘方法研究[J];管理科學學報;2009年02期

3 張紫瓊;葉強;李一軍;;互聯(lián)網(wǎng)商品評論情感分析研究綜述[J];管理科學學報;2010年06期

4 王素格;楊安娜;李德玉;;基于漢語情感詞表的句子情感傾向分類研究[J];計算機工程與應用;2009年24期

5 方美玉;鄭小林;陳德人;華藝;施艷;;商品評論聚焦爬蟲算法設計與實現(xiàn)[J];吉林大學學報(工學版);2012年S1期

6 裴珊珊;葉小梁;;國外Open API發(fā)展現(xiàn)狀及趨勢研究[J];情報科學;2009年12期

本文編號：2080185