本文選題：口碑排序 + 對(duì)比投票��； 參考：《運(yùn)籌與管理》2017年11期

【摘要】：電子商務(wù)時(shí)代,產(chǎn)品的網(wǎng)絡(luò)口碑已成為消費(fèi)者做購(gòu)買決策的重要參考依據(jù)。本文利用客戶評(píng)論信息,依據(jù)產(chǎn)品的評(píng)分及產(chǎn)品之間的對(duì)比投票數(shù)據(jù),提出了一個(gè)新的產(chǎn)品口碑排序方法。首先利用產(chǎn)品兩兩間的對(duì)比投票計(jì)算各自的相對(duì)口碑,再采用貝葉斯平均方法修正原始客戶評(píng)分,然后將二者結(jié)合得到產(chǎn)品的總口碑,進(jìn)而對(duì)產(chǎn)品的網(wǎng)絡(luò)口碑進(jìn)行排序。實(shí)驗(yàn)數(shù)據(jù)采自第三方點(diǎn)評(píng)網(wǎng)站中的產(chǎn)品對(duì)比投票數(shù)據(jù),實(shí)驗(yàn)結(jié)果表明本文提出的產(chǎn)品口碑排序方法具有較高的支持度,且與產(chǎn)品銷量排序的相關(guān)性也很高。
[Abstract]:In the era of e-commerce, the online word-of-mouth of products has become an important reference for consumers to make purchase decisions.In this paper, a new method of product word-of-mouth ranking is proposed based on the customer comment information and the product score and voting data.First, the relative word-of-mouth is calculated by comparing voting between two products, then the original customer score is revised by Bayesian average method, then the total word-of-mouth of the product is obtained by combining the two methods, and then the network word-of-mouth of the product is sorted.The experimental data are collected from the product comparison voting data of the third-party review website. The experimental results show that the proposed method has a high degree of support and has a high correlation with the ranking of product sales.
【作者單位】： 大連理工大學(xué)系統(tǒng)工程研究所;
【基金】：國(guó)家自然科學(xué)基金項(xiàng)目(71471028,71671024) 教育部人文社科項(xiàng)目(15YJCZH198) 遼寧省社科規(guī)劃基金重點(diǎn)項(xiàng)目(L17AGL012) 大連市社科聯(lián)重點(diǎn)項(xiàng)目(2016dlskzd007) 中央高�；究蒲袠I(yè)務(wù)費(fèi)(DUT17JC37)
【分類號(hào)】：C931.1

【相似文獻(xiàn)】

相關(guān)期刊論文 前3條

1 陳宏，鄭家祥;一種新的投資傾斜排序方法──兼與陳國(guó)宏同志商榷[J];管理工程學(xué)報(bào);1996年02期

2 王應(yīng)明，傅國(guó)偉;判斷矩陣排序的X~2方法[J];管理工程學(xué)報(bào);1994年01期

3 王應(yīng)明;群組AHP排序的幾何最小二乘方法[J];中國(guó)管理科學(xué);1996年03期

相關(guān)會(huì)議論文 前1條

1 查勇;梁j;楊鋒;茍清龍;;基于公共權(quán)重的集成DEA排序方法[A];第九屆中國(guó)管理科學(xué)學(xué)術(shù)年會(huì)論文集[C];2007年

，

本文編號(hào)：1739913