Brigitte Grau

Filtering and clustering relations for unsupervised information extraction in open domain

Information Extraction has recently been extended to new areas by loosening the constraints on the strict definition of the extracted information and allowing to design more open information extraction systems. In this new domain of unsupervised information extraction, we focus on the task of extracting and characterizing a priori unknown relations between a given set of entity types. One of the challenges of this task is to deal with the large amount of candidate relations when extracting them from a large corpus. We propose in this paper an approach for the filtering of such candidate relations based on heuristics and machine learning models. More precisely, we show that the best model for achieving this task is a Conditional Random Field model according to evaluations performed on a manually annotated corpus of about one thousand relations. We also tackle the problem of identifying semantically similar relations by clustering large sets of them. Such clustering is achieved by combining a classical clustering algorithm and a method for the efficient identification of highly similar relation pairs. Finally, we evaluate the impact of our filtering of relations on this semantic clustering with both internal measures and external measures. Results show that the filtering procedure doubles the recall of the clustering while keeping the same precision.

Thematic Segmentation of Texts: Two Methods for Two Kind of Texts

To segment texts in thematic units, we present here how a basic principle relying on word distribution can be applied on different kind of texts. We start from an existing method well adapted for scientific texts, and we propose its adaptation to other kinds of texts by using semantic links between words. These relations are found in a lexical network, automatically built from a large corpus. We will compare their results and give criteria to choose the more suitable method according to text characteristics.

Lexical validation of answers in Question Answering

Question answering (QA) aims at retrieving precise information from a large collection of documents, typically the Web. Different techniques can be used to find relevant information, and to compare these techniques, it is important to evaluate question answering systems. The objective of an Answer Validation task is to estimate the correctness of an answer returned by a QA system for a question, according to the text snippet given to support it. In this article, we present a lexical strategy for deciding if the snippets justify the answers, based on our own question answering system. We discuss our results, and show the possible extensions of our strategy.

Selecting Answers to Questions from Web Documents by a Robust Validation Process

Question answering (QA) systems aim at finding answers to question posed in natural language using a collection of documents. When the collection is extracted from the Web, the structure and style of the texts are quite different from those of newspaper articles. We developed a QA system based on an answer validation process able to handle Web specificity. A large number of candidate answers are extracted from short passages in order to be validated according to question and passages characteristics. The validation module is based on a machine learning approach. It takes into account criteria characterizing both passage and answer relevance at surface, lexical, syntactic and semantic levels to deal with different types of texts. We present and compare results obtained for factual questions posed on a Web and on a newspaper collection. We show that our system outperforms a baseline by up to 48% in MRR.

Terminological variants for document selection and question/answer matching

Answering precise questions requires applying Natural Language techniques in order to locate the answers inside retrieved documents. The QALC system, presented in this paper, participated to the Question Answering track of the TREC8 and TREC9 evaluations. QALC exploits an analysis of documents based on the search for multi-word terms and their variations. These indexes are used to select a minimal number of documents to be processed and to give indices when comparing question and sentence representations. This comparison also takes advantage of a question analysis module and recognition of numeric and named entities in the documents.

REGAL, un système pour la visualisation sélective de documents

Information retrieval systems generally return a list of ranked documents, such as only the title and possibly a snippet that contains the words of the request allow a user to evaluate the document relevance relative to her initial request. This kind of result leads the user to browse a lot of documents before satisfying her information need. In order to improve information retrieval, we have studied text visualization: which information has to be shown and how? Our system REGAL (REsume Guide par les Attentes du Lecteur), automatically extracts the visualized information from texts by applying a thematic analysis that does not require a pre-existing structuring or a formatting of the texts, and is based on the combination of two criteria: lexical cohesion and cue phrases. MOTS-CLES : visualisation de texte, navigation textuelle, resume dynamique, analyse thematique.

Evaluation and improvement of cross-lingual question answering strategies

This article presents a bilingual question answering system, which is able to process questions and documents both in French and in English. Two cross-lingual strategies are described and evaluated. First, we study the contribution of biterms translation, and the influence of the completion of the translation dictionaries. Then, we propose a strategy for transferring the question analysis from one language to the other, and we study its influence on the performance of our system.

From Natural Language Requirements to Formal Specification Using an Ontology

In order to check requirement specifications written in natural language, we have chosen to model domain knowledge through an ontology and to formally represent user requirements by its population. Our approach of ontology population focuses on instance property identification from texts. We do so using extraction rules automatically acquired from a training corpus and a bootstrapping terminology. These rules aim at identifying instance property mentions represented by triples of terms, using lexical, syntactic and semantic levels of analysis. They are generated from recurrent syntactic paths between terms denoting instances of concepts and properties. We show how focusing on instance property identification allows us to precisely identify concept instances explicitly or implicitly mentioned in texts.

Validation du type de la réponse dans un système de questions réponses

Les systemes de questions reponses recherchent la reponse a une question posee en langue naturelle dans un ensemble de documents. Certaines questions attendent une reponse d’un certain type, explicite dans la question. La methode presentee dans cet article verifie que la reponse renvoyee correspond bien au type cherche. Pour cela elle suit une approche par apprentissage automatique en utilisant trois types de criteres. Les premiers sont statistiques et fondes sur la frequence d’apparition de la reponse avec le type dans un ensemble de documents. Les seconds relevent de la reconnaissance des entites nommees et les derniers utilisent l’encyclopedie Wikipedia. L’evaluation globale, 80 % de resultats corrects, montre l’interet de la methode.

Towards an Automatic Validation of Answers in Question Answering

Question answering (QA) aims at retrieving precise information from a large collection of documents. Different techniques can be used to find relevant information, and to compare these techniques, it is important to evaluate QA systems. The objective of an Answer Validation task is thus to judge the correctness of an answer returned by a QA system for a question, according to the text snippet given to support it. We participated in such a task in 2006. In this article, we present our strategy for deciding if the snippets justify the answers: a strategy based on our own QA system, comparing the answers it returned with the answer to judge. We discuss our results, then we point out the difficulties of this task.