本文選題：隱式反饋 + 個性化推薦�。� 參考：《北京郵電大學》2015年碩士論文

【摘要】：隨著移動互聯(lián)網(wǎng)的飛速發(fā)展,移動應(yīng)用的日益增多,應(yīng)用市場競爭日益激烈。面對大量的移動應(yīng)用,移動用戶需要花費大量的時間來選擇他們感興趣的應(yīng)用,開發(fā)者也需要花費高昂的推廣成本來獲取用戶。為了便于用戶更方便的找到想要的應(yīng)用,并且能夠提供更多的應(yīng)用曝光機會來幫助移動應(yīng)用開發(fā)者推廣應(yīng)用,降低推廣成本,移動個性化應(yīng)用推薦系統(tǒng)應(yīng)運而生。 本文在傳統(tǒng)推薦系統(tǒng)的基礎(chǔ)上,結(jié)合了移動互聯(lián)網(wǎng)的特性,設(shè)計并實現(xiàn)了一個適用于移動互聯(lián)網(wǎng)的應(yīng)用推薦系統(tǒng)。本文首先介紹了課題背景、研究內(nèi)容以及移動個性化應(yīng)用推薦系統(tǒng)使用到的相關(guān)的技術(shù)。其次結(jié)合了移動互聯(lián)網(wǎng)的特點,針對移動終端用戶和移動應(yīng)用開發(fā)者進行了需求分析,并提出了一種交叉互推的應(yīng)用推薦方式。接著本文根據(jù)移動終端用戶的使用行為和使用習慣,針對傳統(tǒng)Slope One推薦算法的不足,提出了一種基于用戶隱式反饋的混合推薦算法,并通過測試證明混合推薦算法具有更高的準確性。之后基于當前主流移動終端操作系統(tǒng)的特性,提出了適用于移動個性化應(yīng)用推薦系統(tǒng)的用戶標識方法,并根據(jù)對移動用戶和移動開發(fā)者的需求分析,基于用戶隱式反饋的混合推薦算法設(shè)計了具有正確性、可用性以及可擴展性的移動個性化應(yīng)用推薦系統(tǒng),實現(xiàn)了推薦引擎模塊、日志收集處理模塊以及離線計算模塊。最后對實現(xiàn)的系統(tǒng)進行了功能測試和性能測試,并對本文的工作進行了總結(jié)和展望。
[Abstract]:With the rapid development of mobile Internet, the increasing number of mobile applications, application market competition is becoming increasingly fierce. In the face of a large number of mobile applications, mobile users need to spend a lot of time to choose the application they are interested in, and developers also need to spend high cost to get users. In order to facilitate users to find the application they want, and to provide more application exposure opportunities to help mobile application developers promote the application, reduce the promotion costs, mobile personalized application recommendation system came into being. Based on the traditional recommendation system and the characteristics of mobile Internet, this paper designs and implements an application recommendation system for mobile Internet. This paper first introduces the background, research content and the relevant technologies used in mobile personalized application recommendation system. Secondly, combining the characteristics of mobile Internet, this paper analyzes the requirements of mobile terminal users and mobile application developers, and proposes a cross-push application recommendation method. Then, according to the usage behavior and usage habits of mobile terminal users and the shortcomings of traditional Slope One recommendation algorithm, a hybrid recommendation algorithm based on user implicit feedback is proposed in this paper. The test results show that the hybrid recommendation algorithm has higher accuracy. Then, based on the characteristics of the current mainstream mobile terminal operating system, a user identification method suitable for mobile personalized application recommendation system is proposed, and according to the needs of mobile users and mobile developers, Based on the user implicit feedback hybrid recommendation algorithm, a mobile personalized application recommendation system with correctness, availability and extensibility is designed. The recommendation engine module, log collection and processing module and off-line computing module are implemented. Finally, the function test and performance test of the system are carried out, and the work of this paper is summarized and prospected.
【學位授予單位】：北京郵電大學
【學位級別】：碩士
【學位授予年份】：2015
【分類號】：TP391.3

【參考文獻】

相關(guān)期刊論文 前4條

1 蔡登;盧增祥;李衍達;;信息協(xié)同過濾[J];計算機科學;2002年06期

2 黃創(chuàng)光;印鑒;汪靜;劉玉葆;王甲海;;不確定近鄰的協(xié)同過濾推薦算法[J];計算機學報;2010年08期

3 鄧愛林,朱揚勇,施伯樂;基于項目評分預(yù)測的協(xié)同過濾推薦算法[J];軟件學報;2003年09期

4 孟祥武;胡勛;王立才;張玉潔;;移動推薦系統(tǒng)及其應(yīng)用[J];軟件學報;2013年01期

，

本文編號：1782650