This paper describes the work of the Information Retrieval Lab at the University of Delaware (team name “udel”) on TREC 2014 tracks. We participated in five different tracks: Contextual Suggestion, Federated Web, Microblog, Session, and Web.

In many internet applications, such as web searches, contexts are important because each search is done with a specific context in mind, and results outside that context are seen by the user as irrelevant. By measuring joint keyword occurrences in web pages, around our notion of “Semantic Contexts” we propose an infrastructure for semantically characterizing contextual information in the intern...

This paper presents a call-by-need polymorphically typed lambda-calculus with letrec, case, constructors and seq. The typing of the calculus is modelled in a system-F style. Contextual equivalence is used as semantics of expressions. We also define a call-by-name variant without letrec. We adapt several tools and criteria for recognizing correct program transformations to polymorphic typing, in...

Abstract There is growing concern that decision-making informed by machine learning (ML) algorithms may unfairly discriminate based on personal demographic attributes, such as race and gender. Scholars have responded introducing numerous mathematical definitions of fairness to test the algorithm, many which are in conflict with one another. However, these reductionist representations often bear...

A safety case is a hierarchical argument supported by evidence, whose scope defined contextual information. The goal to show that the conclusion of such argument, typically “the system acceptably safe”, true. However, because knowledge about systems always imperfect, value true cannot be assigned with absolute certainty. Instead, researchers have proposed assess belief true, which should high f...

In a pervasive environment, prior knowledge of tasks is not always possible owing to the characteristic uncertainty of the tasks. Moreover, we may not be able to define any tasktemplate at all that can be modeled as a goal for a service composition process. In this paper, we have modeled a service composition process as an event-handling process in the domain of pervasive computing. We have als...

AbstractSentiments of words can differ from one corpus to another. Inducing general sentiment lexicons for languages and using them cannot, in general, produce meaningful results different domains. In this paper, we combine contextual supervised information with the semantic representations occurring dictionary. Contexts help us capture domain-specific scores are indicative polarities those wor...

We introduce a method for content-based advertisement selection for personal blog pages, based on combining multiple representations of the blog. The core idea behind the method is that personal blogs represent individuals, whose interests can be modeled by the language used in the blog itself combined with the language used in related sources of information, such as comments posted to a blog p...

This special issue of Environmental Hazards provides a critical look at the contested relationship between formalised disaster management organisations and emergent power unaffiliated informal volunteers. Using international case studies, five papers identify conceptual, contextual practical challenges opportunities. Much attention is given to definitions; these are important as through definit...