Web Recommender Systems Research Articles

NEWER: A system for NEuro-fuzzy WEb Recommendation

In the era of the Web, there is urgent need for developing systems able to personalize the online experience of Web users on the basis of their needs. Web recommendation is a promising technology that attempts to predict the interests of Web users, by providing them with information and/or services that they need without explicitly asking for them. In this paper we propose NEWER, a usage-based Web recommendation system that exploits the potential of Computational Intelligence techniques to dynamically suggest interesting pages to users according to their preferences. NEWER employs a neuro-fuzzy approach in order to determine categories of users sharing similar interests and to discover a recommendation model as a set of fuzzy rules expressing the associations between user categories and relevances of pages. The discovered model is used by a online recommendation module to determine the list of links judged relevant for users. The results obtained on both synthetic and real-world data show that NEWER is effective for recommendation, leading to a quality of the generated recommendations comparable and often significantly better than those of other approaches employed for the comparison.

Predictors of user perceptions of web recommender systems: How the basis for generating experience and search product recommendations affects user responses

One critical question suggested by Web 2.0 is as follows: When is it better to leverage the knowledge of other users vs. rely on the product characteristic-based metrics for online product recommenders? Three recent and notable changes of recommender systems have been as follows: (1) a shift from characteristic-based recommendation algorithms to social-based recommendation algorithms; (2) an increase in the number of dimensions on which algorithms are based; and (3) availability of products that cannot be examined for quality before purchase. The combination of these elements is affecting users’ perceptions and attitudes regarding recommender systems and the products recommended by them, but the psychological effects of these trends remain unexplored. The current study empirically examines the effects of these elements, using a 2 (recommendation approach: content-based vs. collaborative-based, within)×2 (dimensions used to generate recommendations: 6 vs. 30, between)×2 (product type: experience products (fragrances) vs. search products (rugs), between) Web-based study ( N=80). Participants were told that they would use two recommender systems distinguished by recommendation approach (in fact, the recommendations were identical). There were no substantive main effects, but all three variables exhibited two-way interactions, indicating that design strategies must be grounded in a multi-dimensional understanding of these variables. The implications of this research for the psychology and design of recommender systems are presented.

MUADDIB

Web recommender systems are Web applications capable of generating useful suggestions for visitors of Internet sites. However, in the case of large user communities and in presence of a high number of Web sites, these tasks are computationally onerous, even more if the client software runs on devices with limited resources. Moreover, the quality of the recommendations strictly depends on how the recommendation algorithm takes into account the currently used device. Some approaches proposed in the literature provide multidimensional recommendations considering, besides items and users, also the exploited device. However, these systems do not efficiently perform, since they assign to either the client or the server the arduous cost of computing recommendations. In this article, we argue that a fully distributed organization is a suitable solution to improve the efficiency of multidimensional recommender systems. In order to address these issues, we propose a novel distributed architecture, called MUADDIB, where each user's device is provided with a device assistant that autonomously retrieves information about the user's behavior. Moreover, a single profiler , associated with the user, periodically collects information coming from the different user's device assistants to construct a global user's profile. In order to generate recommendations, a recommender precomputes data provided by the profilers. This way, the site manager has only the task of suitably presenting the content of the site, while the computation of the recommendations is assigned to the other distributed components. Some experiments conducted on real data and using some well-known metrics show that the system works more effectively and efficiently than other device-based distributed recommenders.

Combination of Web page recommender systems

With the rapid growth of the World Wide Web (www), finding useful information from the Internet has become a critical issue. Web recommender systems help users make decisions in this complex information space where the volume of information available to them is huge. Recently, a number of Web page recommender systems have been developed to extract the user behavior from the user’s navigational path and predict the next request as s/he visits Web pages. However, each of these systems has its own merits and limitations. In this paper, we investigate a hybrid recommender system, which combines the results of several recommender techniques based on Web usage mining. We conduct a detailed comparative evaluation of how different combined methods and different recommendation techniques affect the prediction accuracy of the hybrid recommender. We then discuss the results in terms of using a hybrid recommender system instead of a single recommender model. Our results suggest that the hybrid recommender system is better in predicting the next request of a Web user.

Multiobjective evolutionary clustering of Web user sessions: a case study in Web page recommendation

In this study, we experiment with several multiobjective evolutionary algorithms to determine a suitable approach for clustering Web user sessions, which consist of sequences of Web pages visited by the users. Our experimental results show that the multiobjective evolutionary algorithm-based approaches are successful for sequence clustering. We look at a commonly used cluster validity index to verify our findings. The results for this index indicate that the clustering solutions are of high quality. As a case study, the obtained clusters are then used in a Web recommender system for representing usage patterns. As a result of the experiments, we see that these approaches can successfully be applied for generating clustering solutions that lead to a high recommendation accuracy in the recommender model we used in this paper.

Increasing the authoritativeness of web recommendations using PageRank‐based approaches

PurposeWeb content has been widely used for recommending personal webpages. Despite its popularity, the content‐based approach regards a webpage simply as a piece of text, thereby often resulting in less authoritative recommendations of webpages. This paper aims to propose novel approaches that utilise other sources of information pertaining to webpages to facilitate the automatic construction of an authoritative web recommender system.Design/methodology/approachIn this research, four approaches that exploit hyperlink structure, web content and web‐usage logs for making recommendations are proposed. The proposed approaches have been implemented as a prototype system, called the authoritative web recommender (AWR) system. An evaluation using the web‐usage logs and the corresponding pages of a university web site was performed.FindingsThe results from the evaluations using empirical data demonstrate that the four proposed approaches outperform the traditional content‐only approach.Originality/valueThis paper describes a novel way to combine information retrieval, usage mining and hyperlink structure analysis techniques to find relevant and authoritative webpages for recommendation.

A Web Recommendation System using Grid based Support Vector Machines

Main goal of web recommendation system is to study how user behavior on a website can be predicted by analyzing web log data which contain the visited web pages. Many researches of the web recommendation system have been studied. To construct web recommendation system, web mining is needed. Especially, web usage analysis of web mining is a tool for recommendation model. In this paper, we propose web recommendation system using grid based support vector machines for improvement of web recommendation system. To verify the performance of our system, we make experiments using the data set from our web server.

Using the moving average rule in a dynamic web recommendation system

In this, the Information Age, most people are accustomed to gleaning information from the World Wide Web. To survive and prosper, a Web site has to constantly enliven its content while providing various and extensive information services to attract users. The Web Recommendation System, a personalized information filter, prompts users to visit a Web site and browse at a deeper level. In general, most of the recommendation systems use large browsing logs to identify and predict users' surfing habits. The process of pattern discovery is time-consuming, and the result is static. Such systems do not satisfy the end users' goal-oriented and dynamic demands. Accordingly, a pressing need for an adaptive recommendation system comes into play. This article proposes a novel Web recommendation system framework, based on the Moving Average Rule, which can respond to new navigation trends and dynamically adapts recommendations for users with suitable suggestions through hyperlinks. The framework provides Web site administrators with various methods to generate recommendations. It also responds to new Web trends, including Web pages that have been updated but have not yet been integrated into regular browsing patterns. Ultimately, this research enables Web sites with dynamic intelligence to effectively tailor users' needs. © 2007 Wiley Periodicals, Inc. Int J Int Syst 22: 621–639, 2007.

A domain model of Web recommender systems based on usage mining and collaborative filtering

Considering the increasing demand of multi-agent systems, the practice of software reuse is essential to the development of such systems. Multi-agent domain engineering is a process for the construction of domain-specific agent-based reusable software artifacts, like domain models, representing the requirements of a family of multi-agent systems in a domain, and frameworks, implementing reusable agent-based design solutions to those requirements. This article describes the domain modeling tasks of the MADEM methodology and a case study on the application of GRAMO, a MADEM technique, for the construction of the domain model of ONTOWUM, specifying the common and variable requirements of a family of Web recommender systems based on usage mining and collaborative filtering.

A web usage lattice based mining approach for intelligent web personalization

With the explosive growth of information available on the World Wide Web, it has become much more difficult to access relevant information from the Web. One possible approach to solve this problem is web personalization. In this paper, we propose a novel WUL (Web Usage Lattice) based mining approach for mining association access pattern rules for personalized web recommendations. The proposed approach aims to mine a reduced set of effective association pattern rules for enhancing the online performance of web recommendations. We have incorporated the proposed approach into a personalized web recommender system known as AWARS. The performance of the proposed approach is evaluated based on the efficiency and the quality. In the efficiency evaluation, we measure the number of generated rules and the runtime for online recommendations. In the quality evaluation, we measure the quality of the recommendation service based on precision, satisfactory and applicability. This paper will discuss the proposed WUL‐based mining approach, and give the performance of the proposed approach in comparison with the Apriori‐based algorithms.

Hybrid Intelligent Web Recommendation Systems Based on Web Data Mining and Case-Based Reasoning

In this research, we suggest a hybrid intelligent Web recommendation systems based on Web data mining and case-based reasoning (CBR). One of the important research topics in the field of Internet business is blending artificial intelligence (AI) techniques with knowledge discovering in database (KDD) or data mining (DM). Data mining is used as an efficient mechanism in reasoning for association knowledge between goods and customers＇ preference. In the field of data mining, the features, called attributes, are often selected primary for mining the association knowledge between related products. Therefore, most of researches, in the arena of Web data mining, used association rules extraction mechanism. However, association rules extraction mechanism has a potential limitation in flexibility of reasoning. If there are some goods, which were not retrieved by association rules-based reasoning, we can＇t present more information to customer. To overcome this limitation case, we combined CBR with Web data mining. CBR is one of the AI techniques and used in problems for which it is difficult to solve with logical (association) rules. A Web-log data gathered in real-world Web shopping mall was given to illustrate the quality of the proposed hybrid recommendation mechanism. This Web shopping mall deals with remote-controlled plastic models such as remote-controlled car, yacht, airplane, and helicopter. The experimental results showed that our hybrid recommendation mechanism could reflect both association knowledge and implicit human knowledge extracted from cases in Web databases.

Efficient similarity search for market basket data

Several organizations have developed very large market basket databases for the maintenance of customer transactions. New applications, e.g., Web recommendation systems, present the requirement for processing similarity queries in market basket databases. In this paper, we propose a novel scheme for similarity search queries in basket data. We develop a new representation method, which, in contrast to existing approaches, is proven to provide correct results. New algorithms are proposed for the processing of similarity queries. Extensive experimental results, for a variety of factors, illustrate the superiority of the proposed scheme over the state-of-the-art method.