English  |  正體中文  |  简体中文  |  Post-Print筆數 : 27 |  Items with full text/Total items : 113656/144643 (79%)
Visitors : 51718921      Online Users : 655
RC Version 6.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
Scope Tips:
  • please add "double quotation mark" for query phrases to get precise results
  • please goto advance search for comprehansive author search
  • Adv. Search
    HomeLoginUploadHelpAboutAdminister Goto mobile version
    Please use this identifier to cite or link to this item: https://nccur.lib.nccu.edu.tw/handle/140.119/84758


    Title: 整合社群資訊於學習排序模型之推薦系統
    Other Titles: Integrating Social Information by Learning to Rank for Recommendation Systems
    Authors: 蔡銘峰
    Contributors: 資訊科學系 
    Keywords: 資訊檢索;機器學習;自然語言處理;社群網路與分析
    Information Retrieval;Machine Learning;Natural Language Processing;Social Network and Analysis
    Date: 2012
    Issue Date: 2016-04-15 11:34:55 (UTC+8)
    Abstract: 近年來由於網站上社群媒體的盛行,導致於網路上的資訊量迅速增加,因此 如何有效率地找尋資訊已經變得十分重要。而社群推薦系統(Social Recommendation System)在現今的社群網路中,也已經變成一個不可或缺的重要工具。如 何有效率地推薦使用者有興趣的資訊,已經成為近年來資訊相關科學研究中,十 分重要的研究議題。推薦系統的研究亦可視為資訊排序(Information Ranking) 研究之分支,其主旨在於如何根據一些面向資訊來建立模型,此模型可以用 來預測使用者對於某些物品的喜好或是等級評定。模型的建立有二種主流的作 法:一、利用協同過濾技術(Collaborative Filtering Approach),如:矩陣分解 (Matrix Factorization),來根據相似的使用者行為過濾資訊;二、利用內容分析 技術(Content-based Approach),如:向量模型(Vector Space Model),來計算 推薦物品間的相似度以進行推薦。根據文獻中記載,協同過濾的作法中,容易面 臨到冷啟動(Cold Start)或是資訊缺乏(Spareness)等問題。 因此,在此計畫中我們希望利用機器學習排序(Learning to Rank)的技術, 來進行推薦系統模型習得的研究。如何利用機器學習技術習得有效排序,近幾年 來一直是國際資訊檢索的重要研究方向。在此學習架構中,可以使得原本在協 同過濾方法中的使用者社群資訊(Social Information)將可被視為是使用者的屬 性或特徵(attribute/feature),因此常見的特徵整合技術將可以被引用進來解決 資訊缺乏(Spareness)的問題,可能可用的技術包括有:成份分析(Component Analysis)、正規化(Normalization)、特徵組合(Feature Combination)等。另 外,如學習的模型若是利用機率模型來建立的話,甚至常見的統計機率Smoothing 技術也可以應用進來解決此問題;此外,於資訊檢索中,常見的擴展模型 (Propagation Model)也可以用來將物品的資訊透過使用者之間的資訊來進行擴 展,如此將可以讓使用者社群資訊以及推薦物品資訊之間妥善整合,以產生出更 有用的特徵來進行排序模型的學習。因此,在第一年的研究中,我們將會進行如 何機器學習排序技術於推薦系統上的研究,並且期望整合物品資訊和使用者社群 資訊的新型態之資訊特徵可以有效改善推薦系統之效能。 在第二年的計畫中,我們期許開發一個線上推薦系統上,此系統中將會用到第 一年中開發的新技術,並會著重在行動應用程式(Mobile App)的推薦。據最近 數據顯示,全世界的行動應用程式數量,已正式突破了100 萬個,而且每個星期 以1.5 萬個的驚人數量持續增加中。因此,使用者如何選擇一個有用的行動應用 程式也變成是個頭痛的問題。在此應用中,我們預計從Facebook 中收集使用者的 社群資訊,並且透過每個App 之間的中介資訊(metadata)來進行二方面資訊的 整合學習推薦系統,因此第一年的研究成果預計將可以被開發在實際的應用上。
    Social networking websites have becoming more and more popular in recent years. On these websites, there are hundreds of millions of active users creating vast information, which has not been available before. Such vast information, therefore, poses a great challenge in terms of information overload. Social Recommendation Systems is mainly to reduce the information overload over social networking websites by presenting the most relevant information to users. Recommendation systems can also be considered a subclass of information ranking systems that aim to use a model built from the characteristics of an item (content-based approaches) or a user’s social information (collaborative filtering approaches) to predict the “rating” or “preference” that the user would give to the item. Most of previous work uses matrix factorization, a typical collaborative filtering technique, to handle users’ social information such as behavior and activity. Such approaches often suffer from the problems like cold start and sparsity. This work attempts to use learning-to-rank techniques to deal with such a preference problem. In the framework of learning to rank, the social information, such as users’ activity, can be treated as attribute/feature; then, some conventional features integration techniques can be employed to deal with the problems of spareness and missing value, which are the major causes of cold start problem in collaborative filtering approaches. The possible feature integration techniques include component analysis, normalization, and combination; for instance, for a user having no clicked items before, we can use his/her social activity like friendship combining some content-based features from items to make a recommendation. Hence, this approach consists of the techniques of information retrieval, natural language processing, and machine learning. Some statistical techniques like smoothing can also be used to deal with the spareness problem if probabilistic models are being used. Therefore, in this work we aim to use learning-to-rank techniques combining feature integration approaches to well deal with social information for improving the performance of social recommendation systems. Furthermore, in the second year of the project, we aim to apply the techniques developed in the first year to build an on-line social recommendation system for mobile apps. According to some statistics, the total number of mobile apps available online have reached 1 million, and the number is still increasing; therefore, how to select a good-quality mobile app becomes troublesome for mobile users. Combining the social information on Facebook, we attempt to use the first-year research to a real application.
    Relation: 計畫編號 NSC101-2221-E004-017
    Data Type: report
    Appears in Collections:[資訊科學系] 國科會研究計畫

    Files in This Item:

    File Description SizeFormat
    101-2221-E004-017.pdf6359KbAdobe PDF2608View/Open


    All items in 政大典藏 are protected by copyright, with all rights reserved.


    社群 sharing

    著作權政策宣告 Copyright Announcement
    1.本網站之數位內容為國立政治大學所收錄之機構典藏,無償提供學術研究與公眾教育等公益性使用,惟仍請適度,合理使用本網站之內容,以尊重著作權人之權益。商業上之利用,則請先取得著作權人之授權。
    The digital content of this website is part of National Chengchi University Institutional Repository. It provides free access to academic research and public education for non-commercial use. Please utilize it in a proper and reasonable manner and respect the rights of copyright owners. For commercial use, please obtain authorization from the copyright owner in advance.

    2.本網站之製作,已盡力防止侵害著作權人之權益,如仍發現本網站之數位內容有侵害著作權人權益情事者,請權利人通知本網站維護人員(nccur@nccu.edu.tw),維護人員將立即採取移除該數位著作等補救措施。
    NCCU Institutional Repository is made to protect the interests of copyright owners. If you believe that any material on the website infringes copyright, please contact our staff(nccur@nccu.edu.tw). We will remove the work from the repository and investigate your claim.
    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - Feedback