Catálogo de publicaciones - libros

Each user accesses a Website with certain interests. The interest can be manifested by the sequence of each Web user access. The access paths of all Web users can be clustered. The effectiveness and efficiency are two problems in clustering algorithms. This paper provides a clustering algorithm for personalized Web recommendation. It is path clustering based on competitive agglomeration (PCCA). The path similarity and the center of a cluster are defined for the proposed algorithm. The algorithm relies on competitive agglomeration to get best cluster numbers automatically. Recommending based on the algorithm doesn’t disturb users and needn’t any registration information. Experiments are performed to compare the proposed algorithm with two other algorithms and the results show that the improvement of recommending performance is significant.

Preference-based search is the problem of finding an item that matches best with a user’s preferences. User studies show that example-based tools for preference-based search can achieve significantly higher accuracy when they are complemented with suggestions chosen to inform users about the available choices.

We discuss the problem of eliciting preferences in example-based tools and present the lookahead principle for generating suggestions. We compare two different implementations of this principle and we analyze logs of real user interactions to evaluate them.

Nowadays, information gathering in Internet is a complex activity and Internet users need systems to assist them to obtain the information required. In an earlier studies [5, 6, 16] we presented different fuzzy linguistic multi-agent models for helping users in their information gathering processes on the Web. In this paper, we present a new fuzzy linguistic multi-agent model to access information on the Web that incorporates the use of fuzzy multi-granular linguistic modeling to improve its user-system interaction and be more user-friendly.

This paper presents a method to model knowledge in creative environments using dynamic fuzzy ontologies. Dynamic fuzzy ontologies are ontologies that evolve in time to adapt to the environment in which they are used, and whose taxonomies and relationships among concepts are enriched with fuzzy weights (i.e., numeric values between 0 and 1). Such cognitive artifacts can provide for higher user awareness in learning environments, as well as for greater creative stimulus for knowledge discovery. This paper gives the definitions of dynamic fuzzy ontologies, the details of how fuzzy values are dynamically assigned to concepts and relations, and presents an experimental evaluation of the proposed approach.

We introduce a novel personalized web service system through mobile devices. By providing only users’ preferred web pages or smaller readable sections, service elements, the problem of the limitation of resource of mobile devices can be solved. In this paper, the preferred service elements are obtained from the statistical preference transactions among web pages for each web site. In computing the preference, we consider the ratio of the length of each web page and users’ staying time on it. Also, our system dynamically provides the personalized web service according to the different three cases such as the beginning stage, the positive feedback, and the negative feedback. In the experimental section, we demonstrate our personalized web service system and show how much the resource of mobile devices can be saved.

A visual query is based on pictorial representation of conceptual entities and operations. One of the most important features used in visual queries is the shape. Despite its intuitive writing, a shape-based visual query usually suffers of a complexity processing related to two major parameters: 1-the imprecise user request, 2-shapes may undergo several types of transformation. Several methods are provided in the literature to assist the user during query writing. On one hand, relevance feedback technique is widely used to rewrite the initial user query. On the other hand, shape transformations are considered by current shape-based retrieval approaches without any user intervention. In this paper, we present a new cooperative approach based on the shape neighborhood concept allowing the user to rewrite a shape-based visual query according to his preferences with high flexibility in terms of including (or excluding) only some shape transformations and of result sorting.

Knowledge representation and extraction techniques can be efficiently used to improve data modeling and IR functionalities of P2P Information Systems, which have recently attracted a lot of attention from industrial and academic researchers. These functionalities can be achieved by pushing semantics in both data and queries, and exploiting the derived expressiveness to improve file sharing primitives and lookup mechanisms made available from first-generation P2P systems. XML-based P2P Information Systems are a more specific and interesting instance of this class of systems, where the overall data domain is composed by very large, Internet-like distributed XML repositories from which users extract useful knowledge manly by means of IR methodologies implemented on the top of XML join queries. This paper focuses on several aspects of XML-based P2P Information Systems, raging from foundations and definitions to knowledge representation and extraction models and algorithms, along with their experimental evaluation. However, the results presented in this paper can also be adapted to deal with any kind of data format (e.g., HTML).

Ubiquitous recommendation systems predict new items of interest for a user, based on predictive relationship discovered between the user and other participants in Ubiquitous Commerce. In this paper, optimal associative neighbor mining, using attributes, for the purpose of improving accuracy and performance in ubiquitous recommendation systems, is proposed. This optimal associative neighbor mining selects the associative users that have similar preferences by extracting the attributes that most affect preferences. The associative user pattern comprising 3-s (groups of associative users composed of 3-users), is grouped through the ARHP algorithm. The approach is empirically evaluated, for comparison with the nearest-neighbor model and k-means clustering, using the MovieLens datasets. This method can solve the large-scale dataset problem without deteriorating accuracy quality.

Each user accesses a Website with certain interest. The interest is associated with his navigation patterns. The interest navigation patterns represent different interest of the users. In this paper, hybrid Markov model is proposed for interest navigation pattern discovery. The novel model is better in prediction overlay rate and prediction correct rate than traditional Markov models. User group interest is also defined in this paper. The probability of user group interest navigation from one page to another is computed by navigation path characteristics and time characteristics. Compared with the previous ones, the results of the experiment show that the performance is improved efficiently by the hybrid Markov model.

We consider the problem of optimizing processing of batches of frequent itemset queries. The problem is a particular case of multiple-query optimization, where the goal is to minimize the total execution time of the set of queries. We propose an algorithm that is a combination of the Mine Merge method, previously proposed for processing of batches of frequent itemset queries, and the Partition algorithm for memory-based frequent itemset mining. The experiments show that the novel approach outperforms the original Mine Merge and sequential processing in majority of cases.