عنوان مقاله
روند مدل سازی ساختار شناختی یا ادراکی کاربر در زمینه بازیابی اطلاعات متنی
فهرست مطالب
چکیده
مقدمه
DOSAM
ارزیابی آزمایش
نتیجه گیری
بخشی از مقاله
آزمایش 1: ارزیابی کلی اثرات جستجوی شخصی
در آزمایش 1، اثرات جستجو کلاً مورد ارزیابی قرار گرفت. این آزمایش طی دو مرحله انجام گرفت، یکی برای جستجوی شخصی بر اساس DOSAM و دیگری برای جستجوی ساده DLPers V2.0.
درجستجوی ساده، از تکنیک های شخصی استفاده نگردید. در هر دور آزمایش، کلاً 112 جستجو از طرف 52 کاربر (حداقل یک جستجو و حداکثر 3 جستجو برای کاربر) انجام گرفت. نتایج جستجو به طور دستی و به صورت مربوط یا نامربوط مورد ارزیابی قرار گرفت.
کلمات کلیدی:
Modeling User’s Cognitive Structure in Contextual Information Retrieval Xuan Tian tianxuan@ruc.edu.cn Xiaoyong Du duyong@ruc.edu.cn He Hu luckh2@163.com Haihua Li lihhjs@yahoo.com.cn School of Information, Renmin University of China Key Laboratory of Data Engineering and Knowledge Engineering, MOE Beijing, 100872, China Abstract In contextual information retrieval, the retrieval of information depends on the time and place of submitting query, history of interaction, task in hand, and many other factors that are not given explicitly but implicitly lie in the interaction and surroundings of searching, namely the context. User’s cognition is one of important contextual factors for understanding his or her personal needs. We propose a model called DOSAM to get user’s individual cognitive structure on domain knowledge. DOSAM is developed from the spreading-activation model of psychology and is established on the domain ontology. The cost analysis of algorithm shows that it is feasible to get cognitive structure by DOSAM. Personalized search experimental results on digital library indicate that DOSAM can help improve the search effectiveness and user’s satisfaction. 1. Introduction Contextual Information Retrieval (CIR) has been brought forward and became one of research focuses in IR. The retrieval of information depends on the time and place of submitting query, history of interaction, task in hand, and many other factors that are not given explicitly but lie implicitly in the interaction and ambient environment, namely the context [7]. CIR tries to capture user’s needs by augmenting the user’s query with contextual information extracted from his or her searching process. For a user, the context within which he or she seeks information consists of cognitive, social and other factors related to his or her tasks, goals and intentions. As far as cognition is concerned, it is involved in the acquisition and the use of knowledge. It consists of internal Cognitive Structure (CS) and Cognitive Behavior (CB) of knowing in brain [8]. CB is closely related to user’s subjective response, such as feedback, experience, browsing response and so on. CS is different from CB. CS depicts a picture of the way in which the contents of cognition are organized in individual brain, namely the individual picture of knowledge. Cognition has been considerably used for reference in IR. For example, CB was used for intelligent information retrieval interaction and personalized search as in [6, 1]; cognitive framework as a whole, including CS or user’s domain knowledge, was taken into consideration in [2]; common knowledge was also widely applied to query expansion on the ground of general ontology such as WordNet1 and ODP2. To the best of our knowledge, how to exploit user’s individual CS to improve search has so far not been well addressed in the previous work. In this paper, we propose a model called DomainOntology-based Spreading-Activation Model(DOSAM) to get user’s individual CS on domain knowledge. DOSAM is developed from spreading-activation model of psychology, and its goal is to model user’s individual CS on domain knowledge. Since spreading-activation model was introduced by Collins and Loftus in 1975 [4], it has been adopted in various fields. In essence, the effectiveness of spreadingactivation model is crucially dependent on the availability of a representative node association map, and on the use of activation rules that can distinguish the useful nodes from the extraneous ones. In DOSAM, we bring special semantics to relationships between two concepts, develop semantic distances based on concrete semantics and introduce activated strength on concepts. The rest of this paper is organized as follows. Section 2 gives the definitions about DOSAM and shows a feasible algorithm to build DOSAM. Section 3 presents the experimental results about personalized search based on DOSAM. Our conclusions and future work are presented in Section 4.
