Three years after 9/11, the Justice Department made the astounding revelation that more than 120,000 hours of potentially valuable terrorismrelated recordings had yet to be transcribed. Clearly, the government’s efforts to obtain such recordings have continued. Yet there is no evidence that the contents of the recorded calls have been analyzed any more efficiently. Perhaps analysis by conventio...

This paper proposes a novel neural machine reading model for open-domain question answering at scale. Existing machine comprehension models typically assume that a short piece of relevant text containing answers is already identified and given to the models, from which the models are designed to extract answers. This assumption, however, is not realistic for building a large-scale open-domain q...

Coreference resolution systems usually attempt to find a suitable antecedent for (almost) every noun phrase. Recent studies, however, show that many definite NPs are not anaphoric. The same claim, obviously, holds for the indefinites as well. In this study we try to learn automatically two classifications, and , relevant for this problem. We use a small training corpus (MUC-7), but also acquire...

Is information technology (IT) a substitute or complement for inventories? This study answers this question by understanding substitution and complementarity of IT and inventories insofar as they impact profitability of a firm. We use data on IT expenditures and inventories of U.S firms and find support for the substitution between IT and overall inventories. Because overall inventories compris...

The natural language community has made impressive progress in evaluation over the last four years . However, as the evaluations become more sophisticated and more ambitious, a fundamental proble m emerges: how to compare results across changing evaluation paradigms . When we change domain , task, and scoring procedures, as has been the case from MUCK-I to MUCK-II to MUC-3, we los e comparabili...

Alzheimer’s disease (AD) is an irremediable disease that has a high social impact and requires further research. However, the current amount of literature is already overwhelming. This paper shows a case study that applies a machine reading approach to understand the overwhelming amount of papers without much prior knowledge on AD. Machine reading enables us to understand the vast amount of AD ...

Machine reading aims to automatically extract knowledge from text. It is a long-standing goal of AI and holds the promise of revolutionizing Web search and other fields. In this paper, we analyze the core challenges of machine reading and show that statistical relational AI is particularly well suited to address these challenges. We then propose a unifying approach to machine reading in which s...

Cooperation in work settings goes through communicative interactions where colleagues need to reach a certain level of mutual understanding for coordinating their actions or finding negotiated decisions. Mutual understanding is complex because of the heterogeneity of the participants which makes the interpretations uncertain and unpredictable. We argue here that chronic cooperation is still mor...

In this paper we present experiments related to the validation of spoken language understanding capabilities in a language and culture training system. In this application, word-level recognition rates are insufficient to characterize how well the system serves its users. We present the results of an annotation exercise that distinguishes instances of non-recognition due to learner error from i...

GALEN (Generalised Architecture for Languages, Encyclopaedias and Nomenclatures in Medicine) started as a research and development project in the Third Framework Programme to develop a semantically sound model of clinical terminology [1]. In the Fourth Framework’s Galen-In-Use project, the model is further expanded in the domain of surgical procedures and put in practice at various sites for di...