- Full text PDF available (10)
Data Set Used
In this paper we describe a rule-based formalism for the analysis and labelling of texts segments. The rules are contextual rewriting rules with a restricted form of negation. They allow to underspecify text segments not considered relevant to a given task and to base decisions upon context. A parser for these rules is presented and consistence and… (More)
We present our work on the identification of opinions and its components: the source, the topic and the message. We describe a rule-based system for which we achieved a recall of 74% and a precision of 94%. Experimentation with machine-learning techniques for the same task is currently underway. 1 Introduction For some tasks in language processing such as… (More)
In this paper we present the main kernel approaches to the problem of relation extraction from unstructured texts. After a brief introduction to the problem and its characterization as a classification task, we present a survey of the methods and techniques used, and the results obtained. We finally suggest some future lines of work, such as the use of… (More)
Current Information Retrieval systems generally search documents using a keywords model, which is often not expressive enough for the user. In this paper we describe some directions for improving an Information Retrieval system by letting the user specify different semantics constraints in her query, using a language based on a simplified version of… (More)
This paper details the method used to augment an epistemic modality corpus (the Bioscope corpus), incorporating results from the lexical and syntactic analysis of its sentences. The features resulting from these analyses were consolidated in a single data structure, that can be used for interactive experimentation on the corpus. Some visualization aids… (More)
In this paper we present an iterative methodology to improve classifier performance by incorporating linguistic knowledge, and propose a way to incorporate domain rules into the learning process. We applied the methodology to the tasks of hedge cue recognition and scope detection and obtained competitive results on a publicly available corpus.