基于会话推荐的个性化系统外文翻译.docx

1、基于会话推荐的个性化系统外文翻译2015 届华北科技学院本 科 毕 业 设 计（论 文）文 献 翻 译姓 名： 学 号： 专业班级： 院（部）： 指导教师： 2015年 6 月 20 日A Personalized System for Conversational RecommendationsCynthia A. Thompson cindics.utah.eduSchool of ComputingUniversity of Utah50 Central Campus Drive, Rm. 3190Salt Lake City, UT 84112 USAMehmet H. Goker m

2、gokerKaidara Software Inc.330 Distel Circle, Suite 150Los Altos, CA 94022 USAPat Langley langleyisle.orgInstitute for the Study of Learning and Expertise2164 Staunton CourtPalo Alto, CA 94306 USAAbstractSearching for and making decisions about information is becoming increasingly difficult as the am

3、ount of information and number of choices increases. Recommendation systems help users find items of interest of a particular type, such as movies or restaurants, but are still somewhat awkward to use. Our solution is to take advantage of the complementary strengths of personalized recommendation sy

4、stems and dialogue systems, creating personalized aides. We present a system the Adaptive Place Advisor that treats item selection as an interactive, conversational process, with the program inquiring about item attributes and the user responding. Individual, long-term user preferences are unobtrusi

5、vely obtained in the course of normal recommendation dialogues and used to direct future conversations with the same user. We present a novel user model that influences both item search and the questions asked during a conversation. We demonstrate the efiectiveness of our system in significantly red

6、ucing the time and number of interactions required to find a satisfactory item, as compared to a control group of users interacting with a non-adaptive version of the system. 1. Introduction and MotivationRecommendation systems help users find and select items (e.g., books, movies, restaurants) from

7、 the huge number available on the web or in other electronic information sources (Burke, 1999; Resnick & Varian, 1997; Burke, Hammond, & Young, 1996). Given a large set of items and a description of the users needs, they present to the user a small set of the items that are well suited to the descri

8、ption. Recent work in recommendation systems includes intelligent aides for filtering and choosing web sites (Eliassi-Rad & Shavlik, 2001), news stories (Ardissono, Goy, Console, & Torre, 2001), TV listings (Cotter & Smyth, 2000), and other information. The users of such systems often have diverse,

9、conflicting needs. Diffierences in personal preferences, social and educational backgrounds, and private or professional interests are pervasive. As a result, it seems desirable to have personalized intelligent systems that process, filter, and display available information in a manner that suits ea

10、ch individual using them. The need for personalization has led to the development of systems that adapt themselves by changing their behavior based on the inferred characteristics of the user interacting with them (Ardissono & Goy, 2000; Ferrario, Waters, & Smyth, 2000; Fiechter & Rogers, 2000; Lang

11、ley, 1999; Rich, 1979). The ability of computers to converse with users in natural language would arguably increase their usefulness and flexibility even further. Research in practical dialogue systems, while still in its infancy, has matured tremendously in recent years (Allen, Byron, Dzikovska, Fe

12、rguson, Galescu, & Stent, 2001; Dybkjfir, Hasida, & Traum, 2000; Maier, Mast, & Luperfoy, 1996). Todays dialogue systems typically focus on helping users complete a specific task, such as planning, information search, event management, or diagnosis. In this paper, we describe a personalized conversa

13、tional recommendation system designed to help users choose an item from a large set all of the same basic type. Our goal is to support conversations that become more efficient for individual users over time. Our system, the Adaptive Place Advisor, aims to help users select a destination (in this cas

14、e, restaurants) that meets their preferences. The Adaptive Place Advisor makes three novel contributions. To our knowledge, this is the first personalized spoken dialogue system for recommendation, and one of the only conversational natural language interfaces that includes a personalized, long-term

15、 user model. Second, it introduces a novel model for acquiring, utilizing, and representing user models. Third, it is used to demonstrate a reduction in the number of system-user interactions and the conversation time needed to find a satisfactory item. The combination of dialogue systems with perso

16、nalized recommendation addresses weaknesses of both approaches. Most dialogue systems react similarly for each user interacting with them, and do not store information gained in one conversation for use in the future. Thus, interactions tend to be tedious and repetitive. By adding a personalized, lo

17、ng-term user model, the quality of these interactions can improve drastically. At the same time, collecting user preferences in recommendation systems often requires form filling or other explicit statements of preferences on the users part, which can be dificult and time consuming. Collecting prefe

18、rences in the course of the dialogue lets the user begin the task of item search immediately. The interaction between conversation and personalized recommendation has also affected our choices for the acquisition, utilization, and representation of user models. The Adaptive Place Advisor learns info

19、rmation about users unobtrusively, in the course of a normal conversation whose purpose is to find a satisfactory item. The system stores this information for use in future conversations with the same individual. Both acquisition and utilization occur not only when items are presented to and chosen

20、by the user, but also during the search for those items. Finally, the systems representation of models goes beyond item preferences to include preferences about both item characteristics and particular values of those characteristics. We believe that these ideas extend to other types of preferences

21、and other types of conversations. In this paper, we describe our work with the Adaptive Place Advisor. We begin by introducing personalized and conversational recommendation systems, presenting our design decisions along the way. In Section 3 we describe the system in detail, while in Section 4 we p

22、resent our experimental evaluation. In Sections 5 and 6 we discuss related and future work, respectively, and in Section 7 we conclude and summarize the paper. 2. Personalized Conversational Recommendation Systems Our research goals are two-fold. First, we want to improve both interaction quality in

23、 recommendation systems and the utility of results returned by making them user adaptive and conversational. Second, we want to improve dialogue system performance by means of personalization. As such, our goals for user modeling difier from those commonly assumed in recommendation systems, such as

24、improving accuracy or related measures like precision and recall. Our goals also difier from that of previous work in user modeling in dialogue systems (Haller & McRoy, 1998; Kobsa & Wahlster, 1989; Carberry, 1990; Kass, 1991), which emphasizes the ability to track the users goals as a dialogue prog

25、resses, but which does not typically maintain models across multiple conversations. Our hypothesis is that improvements in efficiency and effectiveness can be achieved by using an unobtrusively obtained user model to help direct the systems conversational search for items to recommend. Our approach

26、assumes that there is a large database of items from which to choose, and that a reasonably large number of attributes is needed to describe these items. Simpler techniques might suffice for situations where the database is small or items are easy to describe. 2.1 Personalization Personalized user a

27、daptive systems obtain preferences from their interactions with users, keep summaries of these preferences in a user model, and utilize this model to generate customized information or behavior. The goal of this customization is to increase the quality and appropriateness of both the interaction and

28、 the result(s) generated for each user. The user models stored by personalized systems can represent stereotypical users (Chin, 1989; Rich, 1979) or individuals, they can be hand-crafted or learned (e.g., from questionnaires, ratings, or usage traces), and they can contain information about behavior

29、 such as previously selected items, preferences regarding item characteristics (such as location or price), or properties of the users themselves (such as age or occupation) (Kobsa & Wahlster, 1989; Rich, 1979). Also, some systems store user models only for the duration of one interaction with a use

30、r (Carberry, Chu-Carroll, & Elzer, 1999; Smith & Hipp, 1994), whereas others store them over the long term (Rogers, Fiechter, & Langley, 1999; Billsus & Pazzani, 1998). Our approach is to learn probabilistic, long-term, individual user models that contain information about preferences for items and

31、item characteristics. We chose learned models due to the dificulty of devising stereotypes or reasonable initial models for each new domain encountered. We chose probabilistic models because of their flexibility: a single user can exhibit variable behavior and their preferences are relative rather t

32、han absolute. Long-term models are important to allow influence across multiple conversations. Also, as already noticed, difierent users have difierent preferences, so we chose individual models. Finally, preferences about items and item characteristics are needed to influence conversations and retrieval. Once the decision is made to learn models, another design decision relates to the method by which a system collects preferences for subsequent input to the learning algorithm(s). Here we can distinguish between two approaches. The direct feedback approach places the