本文選題：多媒體社交網(wǎng)絡(luò)　切入點(diǎn)：情境分析　出處：《河南科技大學(xué)》2017年碩士論文

【摘要】：隨著多媒體社交網(wǎng)絡(luò)(Multimedia Social Networks,簡(jiǎn)稱MSNs)的快速發(fā)展,用戶和數(shù)字內(nèi)容的“爆炸式”增長(zhǎng),方便了用戶對(duì)數(shù)字內(nèi)容的訪問,同時(shí)也增加了大量的用戶與用戶,用戶與系統(tǒng)間的交互。針對(duì)這種復(fù)雜的交互如何給用戶提供及時(shí)快捷的個(gè)性化服務(wù)成為多媒體社交網(wǎng)絡(luò)研究的一個(gè)挑戰(zhàn)。因此,本課題的研究在情境感知理論的基礎(chǔ)上發(fā)現(xiàn)用戶在多媒體社交網(wǎng)絡(luò)中的頻繁的行為模式,并根據(jù)用戶的這種行為模式設(shè)計(jì)實(shí)現(xiàn)一種基于情境感知的多媒體內(nèi)容推薦方法,使用戶在海量的數(shù)據(jù)中發(fā)現(xiàn)可能感興趣的內(nèi)容。主要研究?jī)?nèi)容及創(chuàng)新之處如下:1.基于情境感知理論建立一種多媒體社交網(wǎng)絡(luò)中用戶頻繁行為模式發(fā)現(xiàn)的方法。在多媒體社交網(wǎng)絡(luò)環(huán)境中,大量用戶可能在不同的群組中,在相應(yīng)的群組中用戶有不同的角色,用戶在群組中的角色可能使用戶產(chǎn)生不同的意圖,所以在社交媒體下,首先結(jié)合社交媒體特有的屬性在Carl K.Chang教授Situ理論的基礎(chǔ)上建立了SocialSitu理論,對(duì)框架中的元素進(jìn)行定義;其次,基于SocialSitu框架并結(jié)合改進(jìn)的GSP(Generalized Sequential Pattern,廣義序列模式)算法設(shè)計(jì)了多媒體社交網(wǎng)絡(luò)中的用戶行為模式發(fā)現(xiàn)方法;最后,通過該方法分析用戶的歷史SocialSitu(t),可以得出用戶在不同意圖下的行為模式,根據(jù)該用戶的不同意圖下的行為模式可以預(yù)測(cè)用戶當(dāng)前行為序列下的意圖。2.基于用戶的行為模式設(shè)計(jì)多媒體社交網(wǎng)絡(luò)中的多媒體內(nèi)容推薦算法。當(dāng)前現(xiàn)有的推薦系統(tǒng)在用戶對(duì)瀏覽的內(nèi)容打分或進(jìn)行其他操作之前,并不知道用戶對(duì)當(dāng)前瀏覽內(nèi)容的喜好程度,并且用戶的喜好可能隨著用戶所處的環(huán)境以及用戶的身份隨時(shí)發(fā)生變化,通常在多媒體社交網(wǎng)絡(luò)中,用戶有自己的評(píng)分習(xí)慣,或用戶的評(píng)分可能很隨意。因此傳統(tǒng)的推薦算法僅利用相似用戶對(duì)內(nèi)容的評(píng)分來預(yù)測(cè)目標(biāo)用戶的評(píng)分是不準(zhǔn)確的,但是用戶對(duì)多媒體內(nèi)容的操作行為卻能真實(shí)的反應(yīng)用戶對(duì)音視頻的喜好程度。所以為了及時(shí)給用戶推薦可能感興趣的內(nèi)容,本課題的研究在SocialSitu理論的基礎(chǔ)上提出了一種基于用戶情境感知的多媒體內(nèi)容推薦算法,根據(jù)目標(biāo)用戶的近鄰集合對(duì)瀏覽內(nèi)容的歷史操作行為及評(píng)分記錄,以及目標(biāo)用戶瀏覽多媒體內(nèi)容時(shí)的意圖預(yù)測(cè)分析其對(duì)當(dāng)前內(nèi)容的喜好程度,對(duì)目標(biāo)用戶推薦潛在的可能感興趣的內(nèi)容,通過實(shí)驗(yàn)分析,該推薦算法在準(zhǔn)確率和召回率方面有很大提高。在多媒體社交網(wǎng)絡(luò)環(huán)境下給用戶提供個(gè)性化的服務(wù)已經(jīng)成為當(dāng)前服務(wù)系統(tǒng)的趨勢(shì),本課題在多媒體社交網(wǎng)絡(luò)環(huán)境下,針對(duì)用戶的歷史行為數(shù)據(jù)分析挖掘用戶的行為模式,從而根據(jù)這些模式分析用戶潛在的喜好。本文的主要貢獻(xiàn)在于:(1)擴(kuò)展和豐富了Situ理論;(2)通過序列化算法得出用戶不同意圖下的行為序列模式,并根據(jù)某個(gè)用戶在各種意圖下的行為序列模式預(yù)測(cè)該用戶在當(dāng)前行為序列下的意圖;(3)設(shè)計(jì)并實(shí)現(xiàn)了基于情境分析的多媒體內(nèi)容推薦算法,可以及時(shí)的感知用戶潛在的喜好,給用戶提供及時(shí)的個(gè)性化推薦。
[Abstract]:Along with the multimedia social networks (Multimedia Social Networks, referred to as MSNs) the rapid development of users and digital content "explosive" growth, convenient for users to access the digital content, but also increase the number of users and users, interaction between users and system. In view of this complex interaction to provide personalized service timely and quickly become a challenge of multimedia social networks. Therefore, this research found that users in the multimedia social networks frequent behavior model based on context awareness theory, and according to the design of the user behavior implementation of a multimedia content recommendation method based on context awareness, users may find. Interested in the content in the massive data. The main research content and innovation are as follows: 1. the establishment of a social network based on the theory of context aware multimedia Method of network user frequent behavior patterns found in multimedia social network environment, a large number of users in different groups, in the corresponding group of users have different roles in user roles in the group of users may have different intentions, so in the social media, based on Carl K.Chang Situ the theory is established on the basis of SocialSitu theory of social media properties peculiar to the definition of the elements in the framework; secondly, combined with the SocialSitu framework based on improved GSP (Generalized Sequential Pattern, generalized sequential pattern) algorithm to design the model of user behavior of multimedia in social network discovery method; finally, through the analysis of users the history of SocialSitu (T), can be obtained in different intentions under the user behavior model, according to the user's intention of different patterns of behavior can predict the user when The intent of the.2. sequence before the behavior of user behavior patterns based on the design of multimedia content of the multimedia social network recommendation algorithm. The current recommendation system before the user to browse the contents of other operations or the user does not know the score, preference current browsing content, and user preferences may with user environment and the identity of the user to change at any time, usually in the multimedia social networks, users have their own scoring habits, or the user may score is random. So the traditional recommendation algorithm only uses the similar user score on the content to predict the target user's score is not accurate, but the operation behavior of users of multimedia content can like the reaction degree of real users of audio and video. So in order to recommend users may be interested in the contents of the research in Soci On the basis of alSitu theory proposes a recommendation algorithm based on user context aware multimedia content, according to the target user to browse the contents of the neighbor sets of historical operation behavior and scoring records, and intent to target users to browse multimedia content analysis and prediction of its preference for the former when the content, the target user recommended potential sense interested in the content, through experimental analysis, the recommendation algorithm is greatly improved in terms of accuracy and recall rate. In the multimedia social network environment to provide users with personalized service service system has become the current trend, this topic in the multimedia social network environment, according to the history of user behavior data mining and analysis of user behavior patterns thus, the potential user preferences according to the model analysis. The main contribution of this paper is: (1) extended and enriched the theory of Situ; (2) by The serialization algorithm behavior patterns of different user intent, and according to the behavior patterns of a user in a variety of intention prediction under the user in the current sequence of behavior intention; (3) the design and implementation of multimedia content recommendation algorithm based on scenario analysis, timely perception of potential user preferences, to provide users with timely personalized recommendation.

【學(xué)位授予單位】：河南科技大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP391.3;TP393.09

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 李穎基,彭宏,鄭啟倫,曾煒;自動(dòng)分層推薦算法[J];計(jì)算機(jī)應(yīng)用;2002年11期

2 徐義峰;徐云青;劉曉平;;一種基于時(shí)間序列性的推薦算法[J];計(jì)算機(jī)系統(tǒng)應(yīng)用;2006年10期

3 余小鵬;;一種基于多層關(guān)聯(lián)規(guī)則的推薦算法研究[J];計(jì)算機(jī)應(yīng)用;2007年06期

4 張海玉;劉志都;楊彩;賈松浩;;基于頁面聚類的推薦算法的改進(jìn)[J];計(jì)算機(jī)應(yīng)用與軟件;2008年09期

5 張立燕;;一種基于用戶事務(wù)模式的推薦算法[J];福建電腦;2009年03期

6 王晗;夏自謙;;基于蟻群算法和瀏覽路徑的推薦算法研究[J];中國(guó)科技信息;2009年07期

7 周珊丹;周興社;王海鵬;倪紅波;張桂英;苗強(qiáng);;智能博物館環(huán)境下的個(gè)性化推薦算法[J];計(jì)算機(jī)工程與應(yīng)用;2010年19期

8 王文;;個(gè)性化推薦算法研究[J];電腦知識(shí)與技術(shù);2010年16期

9 張愷;秦亮曦;寧朝波;李文閣;;改進(jìn)評(píng)價(jià)估計(jì)的混合推薦算法研究[J];微計(jì)算機(jī)信息;2010年36期

10 夏秀峰;代沁;叢麗暉;;用戶顯意識(shí)下的多重態(tài)度個(gè)性化推薦算法[J];計(jì)算機(jī)工程與應(yīng)用;2011年16期

相關(guān)會(huì)議論文 前10條

1 王韜丞;羅喜軍;杜小勇;;基于層次的推薦:一種新的個(gè)性化推薦算法[A];第二十四屆中國(guó)數(shù)據(jù)庫學(xué)術(shù)會(huì)議論文集（技術(shù)報(bào)告篇）[C];2007年

2 唐燦;;基于模糊用戶心理模式的個(gè)性化推薦算法[A];2008年計(jì)算機(jī)應(yīng)用技術(shù)交流會(huì)論文集[C];2008年

3 秦國(guó);杜小勇;;基于用戶層次信息的協(xié)同推薦算法[A];第二十一屆中國(guó)數(shù)據(jù)庫學(xué)術(shù)會(huì)議論文集（技術(shù)報(bào)告篇）[C];2004年

4 周玉妮;鄭會(huì)頌;;基于瀏覽路徑選擇的蟻群推薦算法:用于移動(dòng)商務(wù)個(gè)性化推薦系統(tǒng)[A];社會(huì)經(jīng)濟(jì)發(fā)展轉(zhuǎn)型與系統(tǒng)工程——中國(guó)系統(tǒng)工程學(xué)會(huì)第17屆學(xué)術(shù)年會(huì)論文集[C];2012年

5 蘇日啟;胡皓;汪秉宏;;基于網(wǎng)絡(luò)的含時(shí)推薦算法[A];第五屆全國(guó)復(fù)雜網(wǎng)絡(luò)學(xué)術(shù)會(huì)議論文（摘要）匯集[C];2009年

6 梁莘q，

本文編號(hào)：1674649