面向推薦的用戶興趣擴(kuò)展方法
發(fā)布時(shí)間:2018-10-12 21:26
【摘要】:提出了一種用戶興趣擴(kuò)展的方法以便應(yīng)用于個(gè)性化推薦系統(tǒng),對(duì)用戶的搜索點(diǎn)擊日志和瀏覽器的瀏覽日志進(jìn)行統(tǒng)計(jì),粗略對(duì)用戶興趣建模,從文本相似度、語(yǔ)言模型相關(guān)度、潛在的語(yǔ)義關(guān)聯(lián)關(guān)系三個(gè)方面充分分析用戶興趣方向之間的關(guān)聯(lián)關(guān)系,應(yīng)用社區(qū)發(fā)現(xiàn)思想挖掘關(guān)聯(lián)關(guān)系緊密的興趣群組,并對(duì)用戶興趣在同一群組內(nèi)進(jìn)行適當(dāng)擴(kuò)展。通過(guò)試驗(yàn)結(jié)果分析,可以看出用戶興趣擴(kuò)展對(duì)個(gè)性化推薦點(diǎn)擊率的影響,并使點(diǎn)擊率有近一倍的增長(zhǎng)。
[Abstract]:In this paper, a method of user interest extension is proposed to be applied to personalized recommendation system. The user search click log and browser browse log are counted, and user interest is modeled roughly, which is based on text similarity and language model relevance. The three aspects of latent semantic association relation fully analyze the correlation relationship between user's interest direction, and apply the idea of community discovery to mine closely related interest groups, and extend user's interest in the same group appropriately. Through the analysis of the experimental results, we can see the influence of user interest expansion on the personalized recommendation click rate, and make the click rate nearly double.
【作者單位】: 長(zhǎng)春工程學(xué)院計(jì)算機(jī)技術(shù)與工程學(xué)院;吉林大學(xué)符號(hào)計(jì)算與知識(shí)工程教育部重點(diǎn)實(shí)驗(yàn)室;
【基金】:國(guó)家自然科學(xué)基金資助項(xiàng)目(61602057) 符號(hào)計(jì)算與知識(shí)工程教育部重點(diǎn)實(shí)驗(yàn)室開放基金資助項(xiàng)目(93K172016K13) 吉林省科技廳優(yōu)秀青年人才基金資助項(xiàng)目(20170520059JH) 吉林省教育廳青年基金資助項(xiàng)目(2016311) 廣西可信軟件重點(diǎn)實(shí)驗(yàn)室研究課題資助項(xiàng)目(kx201533)
【分類號(hào)】:TP391.3
[Abstract]:In this paper, a method of user interest extension is proposed to be applied to personalized recommendation system. The user search click log and browser browse log are counted, and user interest is modeled roughly, which is based on text similarity and language model relevance. The three aspects of latent semantic association relation fully analyze the correlation relationship between user's interest direction, and apply the idea of community discovery to mine closely related interest groups, and extend user's interest in the same group appropriately. Through the analysis of the experimental results, we can see the influence of user interest expansion on the personalized recommendation click rate, and make the click rate nearly double.
【作者單位】: 長(zhǎng)春工程學(xué)院計(jì)算機(jī)技術(shù)與工程學(xué)院;吉林大學(xué)符號(hào)計(jì)算與知識(shí)工程教育部重點(diǎn)實(shí)驗(yàn)室;
【基金】:國(guó)家自然科學(xué)基金資助項(xiàng)目(61602057) 符號(hào)計(jì)算與知識(shí)工程教育部重點(diǎn)實(shí)驗(yàn)室開放基金資助項(xiàng)目(93K172016K13) 吉林省科技廳優(yōu)秀青年人才基金資助項(xiàng)目(20170520059JH) 吉林省教育廳青年基金資助項(xiàng)目(2016311) 廣西可信軟件重點(diǎn)實(shí)驗(yàn)室研究課題資助項(xiàng)目(kx201533)
【分類號(hào)】:TP391.3
【相似文獻(xiàn)】
相關(guān)期刊論文 前10條
1 鄭運(yùn)剛;馬建國(guó);;基于分類的用戶興趣漂移模型[J];情報(bào)雜志;2008年01期
2 楊杰;陳恩紅;;面向個(gè)性化服務(wù)的用戶興趣偏移檢測(cè)及處理方法[J];電子技術(shù);2009年11期
3 陳圣兵;李龍澍;紀(jì)霞;;多層次用戶興趣模式的動(dòng)態(tài)捕捉[J];計(jì)算機(jī)工程與應(yīng)用;2009年36期
4 鄭曉健;龐淑英;何英;;一種面向主題的用戶興趣挖掘模型研究[J];昆明學(xué)院學(xué)報(bào);2010年03期
5 花青松;劉海峰;胡錚;;基于基尼系數(shù)的用戶興趣分布模式度量方法[J];計(jì)算機(jī)工程;2012年22期
6 蔣學(xué)鋒;;用戶興趣的結(jié)構(gòu)和個(gè)性化服務(wù)的實(shí)現(xiàn)[J];計(jì)算技術(shù)與自動(dòng)化;2005年04期
7 李鈍;曹元大;張龍飛;;用戶興趣優(yōu)化過(guò)濾方法研究[J];計(jì)算機(jī)工程;2006年20期
8 費(fèi)洪曉;戴弋;穆s,
本文編號(hào):2267652
本文鏈接:http://www.sikaile.net/kejilunwen/ruanjiangongchenglunwen/2267652.html
最近更新
教材專著