Search results

The purpose of this paper is to improve users’ search results relevancy by manipulating their explicit feedback.

CoRRe – an explicit feedback model integrating three popular feedback, namely, Comment-Rating-Referral is proposed in this study. The model is further enhanced using case-based reasoning in retrieving the top-5 results. A search engine prototype was developed using Text REtrieval Conference as the document collection, and results were evaluated at three levels (i.e. top-5, 10 and 15). A user evaluation involving 28 students was administered, focussing on 20 queries.

Both Mean Average Precision and Normalized Discounted Cumulative Gain results indicate CoRRe to have the highest retrieval precisions at all the three levels compared to the other feedback models. Furthermore, independent t-tests showed the precision differences to be significant. Rating was found to be the most popular technique among the participants, producing the best precision compared to referral and comments.

The findings suggest that search retrieval relevance can be significantly improved when users’ explicit feedback are integrated, therefore web-based systems should find ways to manipulate users’ feedback to provide better recommendations or search results to the users.

The study is novel in the sense that users’ comment, rating and referral were taken into consideration to improve their overall search experience.

An important issue in the management of a web‐based user community, where users are registered to a web portal, is to identify patterns of users' interest. In this context, the users' feedback plays a major role. The purpose of this paper is to define a novel framework analysis for managing the feedback given by registered visitors of a web site.

The paper presents a new technique to integrate the feedback explicitly given by users into already existing user profiles. The authors introduce the novel concepts of scope, filtering, and relevance profiles for managing users' feedback. The new concept of Relevance Profile (RP) is defined.

Using the framework, the authors were able to discover patterns of usage of registered users of a web site.

The practical applicability of the approach is validated by a use case study showing how the framework can be used with a real web site. The authors used Gugubarra as a reference system, a prototype for creating and managing web user profiles, developed by the DBIS group at the Goethe‐University of Frankfurt.

A new way to integrate the user feedback into interest profiles and a novel framework to analyze and discover patterns of interests are presented. The paper is an extended version (more than 50 per cent novel material) of a previous paper presented at the iiWAS2010 conference.

Seeking and retrieving information is an essential aspect of knowledge workers' activities during problem‐solving and decision‐making tasks. In recent years, user‐oriented Information Seeking (IS) research methods rooted in the social sciences have been integrated with Information Retrieval (IR) research approaches based on computer science to capitalize on the strengths of each field. Given this background, the objective is to develop a topic‐needs variation determination technique based on the observations of IS&R theories.

In this study, implicit and explicit methods for identifying users' evolving topic‐needs are proposed. Knowledge‐intensive tasks performed by academic researchers are used to evaluate the efficacy of the proposed methods. The paper conducted two sets of experiments to demonstrate and verify the importance of determining changes in topic‐needs during the IS&R process.

The results in terms of precision and discounted cumulated gain (DCG) values show that the proposed Stage‐Topic_W (G,S) and Stage‐Topic‐Interaction methods can retrieve relevant document sets for users engaged in long‐term tasks more efficiently and effectively than traditional methods.

The improved precision of the proposed methods means that they can retrieve more relevant documents for the searcher. Accordingly, the results of this research have implications for enhancing the search function in enterprise content management (ECM) applications to support the execution of projects/tasks by professionals and facilitate effective ECM.

The model observes a user's search behavior pattern to determine the personal factors (e.g. changes in the user's cognitive status), and content factors (e.g. changes in topic‐needs) simultaneously. The objective is to capture changes in the user's information needs precisely so that evolving information needs can be satisfied in a timely manner.

Good abandonment behavior refers to users obtaining direct answers via search engine results pages (SERPs) without clicking any search result, which occurs commonly in mobile search. This study aims to better understand users' good abandonment behavior and perception, and then construct a good abandonment prediction model for mobile search with improved performance.

In this study, an in situ user mobile search experiment (N = 43) and a crowdsourcing survey (N = 1,379) were conducted. Good abandonment behavior was analyzed from a quantitative perspective, exploring users' search behavior characteristics from four aspects: session and query, SERPs, gestures and eye-tracking data.

Users show less engagement with SERPs in good abandonment, spending less time and using fewer gestures, and they pay more visual attention to answer-like results. It was also found that good abandonment behavior is often related to users' perceived difficulty of the searching tasks and trustworthiness in the search engine. A good abandonment prediction model in mobile search was constructed with a high accuracy (97.14%).

This study is the first to explore eye-tracking characteristics of users' good abandonment behavior in mobile search, and to explore users' perception of their good abandonment behavior. Visual attention features are introduced into good abandonment prediction in mobile search for the first time and proved to be important predictors in the proposed model.

In this “Info-plosion” era, recommendation systems (or recommenders) play a significant role in finding interesting items in the surge of online digital activity and e-commerce. The purpose of this paper is to model users' preference evolution to recommend potential items which users may be interested in.

A novel recommendation system, namely evolution-learning recommendation (ELR), is developed to precisely predict user interest for making recommendations. Differing from prior related methods, the authors integrate the matrix factorization (MF) and recurrent neural network (RNN) to effectively describe the variation of user preferences over time.

A novel cumulative factorization technique is proposed to efficiently decompose a rating matrix for discovering latent user preferences. Compared to traditional MF-based methods, the cumulative MF could reduce the utilization of computation resources. Furthermore, the authors depict the significance of long- and short-term effects in the memory cell of RNN for evolution patterns. With the context awareness, a learning model, V-LSTM, is developed to dynamically capture the evolution pattern of user interests. By using a well-trained learning model, the authors predict future user preferences and recommend related items.

Based on the relations among users and items for recommendation, the authors introduce a novel concept, virtual communication, to effectively learn and estimate the correlation among users and items. By incorporating the discovered latent features of users and items in an evolved manner, the proposed ELR model could promote “right” things to “right” users at the “right” time. In addition, several extensive experiments are performed on real datasets and are discussed. Empirical results show that ELR significantly outperforms the prior recommendation models. The proposed ELR exhibits great generalization and robustness in real datasets, including e-commerce, industrial retail and streaming service, with all discussed metrics.

Online businesses require a deep understanding of their customers’ interests to innovate and develop new products and services. Users, on the other hand, rarely express their interests explicitly. The purpose of this study is to predict users’ implicit interest in products of an online store based on their mouse behavior through various product page elements.

First, user mouse behavior data is collected throughout an online store website. Next, several mouse behavioral features on the product pages elements are extracted and finally, several models are extracted using machine learning techniques to predict a user’s interest in a product.

The results indicate that focusing on mouse behavior on various page elements improves user preference prediction accuracy compared to other available methods.

User mouse behavior was used to predict consumer preferences in this study, therefore gathering additional data on user demography, personality dimensions and emotions may significantly aid in accurate prediction.

Mouse behavior is the most repeated behavior during Web page browsing through personal computers and laptops. It has been referred to as implicit feedback in some studies and an effective way to ascertain user preference. In these studies, mouse behavior is only assessed throughout the entire Web page, lacking a focus on different page elements. It is assumed that in online stores, user interaction with key elements of a product page, such as an image gallery, user reviews, a description and features and specifications, can be highly informative and aid in determining the user’s interest in that product.

Tourism spot or point-of-interest (POI) recommendation has become a common service in people's daily life. The purpose of this paper is to model users' check-in history in order to predict a set of locations that a user may soon visit.

The authors proposed a novel learning-based method, the pattern-based dual learning POI recommendation system as a solution to consider users' interests and the uniformity of popular POI patterns when making recommendations. Differing from traditional long short-term memory (LSTM), a new users’ regularity–POIs’ popularity patterns long short-term memory (UP-LSTM) model was developed to concurrently combine the behaviors of a specific user and common users.

The authors introduced the concept of dual learning for POI recommendation. Several performance evaluations were conducted on real-life mobility data sets to demonstrate the effectiveness and practicability of POI recommendations. The metrics such as hit rate, precision, recall and F-measure were used to measure the capability of ranking and precise prediction of the proposed model over all baselines. The experimental results indicated that the proposed UP-LSTM model consistently outperformed the state-of-the-art models in all metrics by a large margin.

This study contributes to the existing literature by incorporating a novel pattern–based technique to analyze how the popularity of POIs affects the next move of a particular user. Also, the authors have proposed an effective fusing scheme to boost the prediction performance in the proposed UP-LSTM model. The experimental results and discussions indicate that the combination of the user's regularity and the POIs’ popularity patterns in PDLRec could significantly enhance the performance of POI recommendation.

The purpose of this paper is not only to design a more effective recommendation system for libraries, but also to eliminate many of the weaknesses found in the existing library recommender systems.

A novel library recommender system was designed for English collections by integrating personal ontology model and collaborative filtering model with domain specification.

The trend of the traditional library is evolving toward that of digital library. The personal ontology recommender (PORE) system offers a friendly user interface and provides several personalized services.

This system is only implemented and tested in the Library of National Chung Hsing University in Taiwan.

The paper demonstrates a good methodology to offer an active, effective, and personalized recommendation system for library patrons.

The purpose of this paper is to describe the development of a personalised information support system to help faculty members to search various portals and e‐resources without typing the search terms in different interfaces and to obtain results re‐ordered without human intervention.

After a careful survey of various tools and techniques available for computerised client‐centred information services, the study selected to apply web usage mining, proxy level data collection and a vector space retrieval model to develop the personalised information support for teaching and research in a higher education institution.

There are practical constraints in the implementation stage. There is considerable difficulty in getting real and correct user interests and mapping them effectively into the products and services offered by the library. Also the interests of users change continuously. If multiple users share the same PC, it is difficult to identify the user as there is no one‐to‐one mapping between user and IP address.

The paper has not considered cases for all the faculty members due to time constraints. The results obtained from the system also need correlation with the sources actually used by the faculty to test its efficacy in a highly fluid research situation like higher academics.

A pragmatic client‐centred information support prototype described in this paper may find use in other institutions needing similar information support.

This paper demonstrates the pragmatic application of ICT for linking users and e‐resources in an academic library.

This study aims to determine how to influence the customer journey of mobile food ordering applications (MFOAs) users. It researches how available information could influence customers’ intention to use MFOAs platforms in the prepurchase stage and explores the potential of personalized information to improve customer satisfaction with these services in the postpurchase stage.

This research followed a mixed design, combining qualitative (focus groups) and quantitative (online survey) research and using both content analysis and partial least squares structural equation modeling.

Two types of available information (firm-generated information and online customer reviews) had a positive influence on the behavioral intention to use MFOAs. Additionally, findings showed that different web personalization strategies, namely, content personalization, functional personalization and system-driven personalization, were useful tools to create customer satisfaction with this type of platform.

The study discusses limitations regarding the sample and sampling process, indicator variables and measures.

The present research provides actionable insights for online food delivery providers.

This study addresses a research gap in the literature and provides a novel and richer understanding of customer behavior toward mobile food delivery platforms. Also, it adds to the personalization research by identifying and testing a range of web personalization strategies.