Search

In domain-specific search, effectiveness depends on considering multiple dimensions of relevance, beyond topicality, related to the domain involved. Estimating additional relevance dimensions can affect efficiency because the computation of relevance scores is time-consuming if performed on the full document, especially for query-dependent dimensions. Hence, this article introduces an approach for improving effectiveness in domain-specific search by considering multiple dimensions of relevance, namely topicality, correctness, and credibility. To address efficiency, we propose a re-ranking approach that estimates domain-specific relevance scores on document summaries, rather than full documents. We validate the proposed solution by performing the Ad-Hoc Retrieval task from the TREC 2020 Health Misinformation Track, a domain that crucially relies on the considered relevance dimensions. Our findings underscore the potential of our approach with respect to both effectiveness and efficiency.

This paper primarily addresses a dataset relating to cellular, chemical and physical conditions of patients gathered at the time they are operated upon to remove colorectal tumours. This data provides a unique insight into the biochemical and immunological status of patients at the point of tumour removal along with information about tumour classification and post-operative survival. The relationship between severity of tumour, based on TNM staging, and survival is still unclear for patients with TNM stage 2 and 3 tumours. We ask whether it is possible to predict survival rate more accurately using a selection of machine learning techniques applied to subsets of data to gain a deeper understanding of the relationships between a patient's biochemical markers and survival. We use a range of feature selection and single classification techniques to predict the 5 year survival rate of TNM stage 2 and 3 patients which initially produces less than ideal results. The performance of each model individually is then compared with subsets of the data where agreement is reached for multiple models. This novel method of selective ensembling demonstrates that significant improvements in model accuracy on an unseen test set can be achieved for patients where agreement between models is achieved. Finally we point at a possible method to identify whether a patients prognosis can be accurately predicted or not.

In recent years there has been a great deal of research about personalization in information access. The main issue is to improve the quality of search by producing user- tailored results related to specific user needs. In this context two key research problems concerns how to model user preferences, and how to exploit them in effective personalisation processes. This talk will address the above problems, and will discuss some important issues related to their possible solutions. The talk will also point out some research directions worth to be explored.

This paper shortly discusses the main issues related to the problem of personalizing search. To overcome the “one size fits all” behavior of most search engines and Information Retrieval Systems, in recent years a great deal of research has addressed the problem of defining techniques aimed at tailoring the search outcome to the user context. This paper outlines the main issues related to the two basic problems beyond these approaches: context representation and definition of processes which exploit the context knowledge to improve the quality of the search outcome. Moreover some other important and related issues are mentioned, such as privacy, and evaluation.

In this paper an original soft hierarchical Fuzzy Clustering algorithm is proposed, named Hierarchical Hyper-spherical Divisive Fuzzy C-Means (H2D-FCM), with the following characteristics: it generates a "soft" hierarchy in which a document can belong to several child clusters of a node, and the clusters in the same hierarchical level are more specific (general) than the clusters in the upper (lower) level. The proposed algorithm is a divisive algorithm based on a modified bisective K-Means, applying a modified probabilistic Fuzzy C Means algorithm to divide each node into child-nodes. The algorithm determines the proper number of cluster to generate at the first level based on an entropy measure and decides if a node can be further split based on a "density" measure. The paper presents the algorithm and its evaluations on two standard collections. © 2009 IEEE.

In the last decade, new approaches focused on modelling uncertainty over complex relational data have been developed. In this paper one of the most promising of such approaches, known as Probabilistic Relational Models (PRMs), has been investigated and extended in order to measure and include uncertainty over relationships. Our extension, called PRMs with Relational Uncertainty, has been evaluated on real-data for web document classification purposes. Experimental results shown the potentiality of the proposed methods of capturing the real "strength" of relationships and the capacity of including this information into the probability model

The proceedings contain 150 papers. The topics discussed include: new techniques for data preprocessing based on usage logs for efficient web user profiling at client side; managing context-dependent workspace awareness in an e-collaboration environment; enhanced gestalt theory guided web page segmentation for mobile browsing; automated web site evaluation - an approach based on ranking SVM; analyzing social networks using FCA: complexity aspects; optimizing web content presentation: a online pso approach; clustering with constrained similarity learning; USE: a concept-based recommendation system to support creative search; user study of automatic photo classifier by color and timestamp; visualization cube: modeling interaction for exploratory data analysis of spatiotemporal trend information; and how does label propagation algorithm work in bipartite networks?

In this paper we present a preliminary analysis of a new Google Desktop gadget aimed at defining personalised query expansions. This user friendly application is based on a dynamic local fuzzy ontology: every time that a user interacts with the gadget the ontology relationships are updated. The proposed query expansion technique takes into account past user's queries and the concepts contained in relevant documents stored into the user's workstation. © 2008 IEEE

Multisource spatial data fusion strategies have generally to cope with distinct kinds of uncertainty, deriving by both the trust of the source, the imperfection of the spatial data, and the vagueness of the fusion strategy itself. In this paper we propose a consensual fusion that allows to flexibly model several fusion strategies ranging from a risk-taking to a risk-adverse attitude, and capable to cope with both data imprecision and uncertainty and source reliability. Uncertainty and imprecision in spatial data is represented by associating a fuzzy value with each spatial unit. The fusion function has a quantifier-guided nature reflecting the concept of a fuzzy majority and works on imprecise values to compute an imprecise result. It is formalized by a generalized OWA operator defined in the paper for aggregating imprecise values with distinct importance. The consensual fusion works so that the greater the trust score of the source and its agreement with the other sources, the more influent (important) is the data from the source in determining the consensual values. Further, when the data are affected by uncertainty one can require to fuse them so as to compute a result affected by at most a given maximum uncertainty level.

The PENG system is intended to provide an integrated and personalized environment for news professionals, providing functionalities for filtering, distributed retrieval, and a,flexible interface environment for the display and manipulation of news materials. In this paper we review the progress and results of the PENG system to date, and describe in detail the document filtering part of the system, which is designed to gather and filter documents to user profiles. The current architecture will be described, along with some of the main issues which have so far been found in it's development.

In this paper an overview of the PENG project and its system architecture is presented. The objective of the project is to define and develop a news content composition and programming environment, to provide news professionals and general users with an interactive and personalised tool for multimedia news gathering and delivery. This is achieved by defining and developing a flexible system for a personalised filtering, retrieval and composition of multimedia news.