Christine Jacquin

Automatic named identification of speakers using diarization and ASR systems

In this paper, we consider the extraction of speaker identity from audio records of broadcast news without a priori acoustic information about speakers. Using an automatic speech recognition system and an automatic speaker diarization system, we present improvements for a method which allows to extract speaker identities from automatic transcripts and to assign them to speech segments. Experiments are carried out on French broadcast news records from the ESTER 1 evaluation campaign. Experimental results using outputs of automatic speech recognition and automatic diarization are presented.

French EuroWordNet Lexical Database Improvements

Semantic knowledge is often used in the framework of Natural Language Processing (NLP) applications. However, for some languages different from English, such knowledge is not always easily available. In fact, for example, French thesaurus are not numerous and are not enough developed. In this context, we present two modifications made on the French version of the EuroWordnet Thesaurus in order to improve it. Firstly, we present the French EuroWordNet thesaurus and its limits. Then we explain two improvements we have made. We add non-existing relationships by using the bilinguism capability of the EuroWordnet thesaurus, and definitions by using an external multilingual resource (Wikipedia [1]).

A Web site indexing process for an Internet information retrieval agent system

Bonom is a distributed agent based system to perform information retrieval on the Internet. The middle agents which perform the matching between site agents (information providers) and user agents (requesters) are numerous and can be world wide distributed. Moreover, the agents are structured into nested communities according to a hierarchy of information domains. This paper focuses on a Web site indexing process which uses disambiguated ontologies. It allows the site agent to know to which communities it has to subscribe to, in order to get requests it can accurately process.

Dinosys: An Annotation Tool for Web-Based Learning

The main function of freeform annotation systems is to improve exchange, communication and interoperability on the Web. The purpose of this paper is on the one hand, to make a synthesis of the characteristics of the annotations and architectures of annotation systems, and on the other hand to propose a new architecture for an annotation system which is easy to use, lightweight, efficient, non-intrusive, scaleable, shared and platform-independent. The use of this tool within the e-learning framework is also studied.

Clustering short text and its evaluation

Recently there has been an increase in interest towards clustering short text because it could be used in many NLP applications. According to the application, a variety of short text could be defined mainly in terms of their length (e.g. sentence, paragraphs) and type (e.g. scientific papers, newspapers). Finding a clustering method that is able to cluster short text in general is difficult. In this paper, we cluster 4 different corpora with different types of text with varying length and evaluate them against the gold standard. Based on these clustering experiments, we show how different similarity measures, clustering algorithms, and cluster evaluation methods effect the resulting clusters. We discuss four existing corpus based similarity methods, Cosine similarity, Latent Semantic Analysis, Short text Vector Space Model, and Kullback-Leibler distance, four well known clustering methods, Complete Link, Single Link, Average Link hierarchical clustering and Spectral clustering, and three evaluation methods, clustering F-measure, adjusted Rand Index, and V. Our experiments show that corpus based similarity measures do not significantly affect the clusters and that the performance of spectral clustering is better than hierarchical clustering. We also show that the values given by the evaluation methods do not always represent the usability of the clusters.

The answer validation system ProdicosAV dedicated to French

In this paper, we present the ProdicosAV answer validation system which was developed by the NLP team from the LINA institute. ProdicosAV system is based on the Prodicos System which participated two years ago in the Question Answering CLEF evaluation campaign for French. We firstly present the modifications made on Prodicos to improve it and to adapt it to a new kind of exercise. We present in details the ranking passage module and the temporal validator module. Secondly, the answer-validation module dedicated to the AVE task is presented. Finally, the evaluation is put forward to justify the results obtained.

The query answering system PRODICOS

In this paper, we present the PRODICOS query answering system which was developed by the TALN team from the LINA institute. We present the various modules constituting our system and for each of them the evaluation is shown. Afterwards, for each of them, the evaluation is put forward to justify the results obtained. Then, we present the main improvement based on the use of semantic data.

Question types specification for the use of specialized patterns in prodicos system

We present the second version of the Prodicos query answering system which was developed by the TALN team from the LINA institute. The main improvements made concern in the one hand, the use of external knowledge (Wikipedia) to improve the passage selection step. And on the other hand, the answer extraction step is improved by the determination of four different strategies for locating the answer to a question regarding its type. Afterwards, for the passage selection and answer extraction modules, the evaluation is put forward to justify the results obtained.

Indexing a Web Site to Highlight Its Content

This article presents a new approach in order to index a Web site. It uses ontologies and natural language techniques for information retrieval on the Internet. The main goal is to build a structured index of the Web site. This structure is given by a terminology oriented ontology of a domain which is chosen a priori according to the content of the Web site. The indexing process uses improved natural language technics.