Chantal Enguehard

Flexible-Equality of Terms: Definition and Evaluation

The recognition of index terms is a very frequent way of answering a query through the identification of documents which fit the query. However, these terms can vary in their form (singular or plural, verbal or nominal form, etc.) and become difficult to identify. We present the flexible-equality of terms which determines whether two terms can be considered as variations one from the other. This operator is based on the minimum editing distance between two strings and has been extended to complex terms (composed of several words). This operator does not require large amounts of linguistic resources other than a list of functional words. We evaluate the performances of the algorithm by comparison with the results of FASTR, a system dedicated to the recognition of complex terms, on an English corpus. This system is based on linguistic treatments and uses linguistic resources. We found a rate of recall of 76,6% and a rate of precision of 96,3% on complex terms.

Indexation de textes: l'apprentissage des concepts

In technical fields, many documents go unread due to a lack of awareness of their existence. A system which indexes texts can find all relevant texts in response to a query. The problem is to establish the indexation. At present, advanced full text systems automatically index texts on the complete thesaurus with computed weights. Another way of doing this can be a person choosing the set of relevant concepts. This second solution is better but more costly and dependent on the classification choices made by the operator.To meet these problems, ANA (Auomatic Natural Acquisition) had been developed. This system automatically extracts relevant concepts from free texts to produce a semantic network. It does not rely on grammar or lexicon but, instead, is based on an original statistical method.This research brings about two developments: on one hand the system is also capable of extracting the simple grammatical structures it encounters, most often in order to improve its performance, and on the other hand this will lead to an automatic definition of semantic classes of concepts, in order to structure the network.

ANA: AUTOMATIC NATURAL ACQUISITION

Any system for natural language processing must be based on a lexicon. Once a model has been defined, there is the problem of acquiring and inserting words. This task is tedious for a human operator; on the one hand he must not forget any of the words, and on the other the acquisition of a new concept requires the input of a number of parameters. In view of these difficulties, research work has been undertaken in order to integrate pre-existing “paper” dictionaries. Nevertheless, these are not faultless, and are often incomplete when processing a very specialized technical field. We have therefore searched to mitigate these problems by automating the enrichment of an already partially integrated lexicon. We work in a technical field on which we have gathered different sorts of texts: written texts, specialist interviews, technical reports, etc. These documents are stored in an object oriented database, and form part of a wider project, called REX (“Retour d’EXperience” in French, or “Feedback of Experience” in English). Our system, called ANA, reads the documents, analyses them, and deduces new knowledge, allowing the enrichment of the lexicon. The group of words already integrated into the lexicon form the “Bootstrap” of the discovery process of new words: it collects the instances of the different concepts thought to be interesting, in order to gather the semantic information. A special module makes it possible to avoid an explosion of the size of the database. It is responsible for forgetting certain instances and maintaining the database in such a way that the order in which the texts are introduced bears no influence.

Conventional medical attitudes to using a traditional medicine vodou-based model of pain management: survey of French dentists and the proposal of a pain model to facilitate integration.

Objectives The purposes of this study were to develop a pain management model using traditional medicine (TM) vodou healing methods; to survey a sample of French dentists to rate components of conventional and proposed TM vodou-based pain management model; and to assess the possibility of conventional, allopathic providers to integrate TM or complementary and alternative medicine concepts. Methods From a set of 30 fact sheets collected from TM African healers (vodou healers), main clinical concepts and terminology were extracted. Twenty vodou-based pain management concepts were collected from an interview with a TM vodou practitioner. From this information, a 7-step vodou-based pain management model was created. A sample of 40 French dentists from Nantes, France, whose practices focused on the clinical treatment of dental pain, was surveyed to assess the importance of both TM (vodou) and conventional biomedical components. Results Seventy percent of the dentists sampled rated the rational components of the TM model as “very important” or “important” for pain treatment, whereas 2 other traditional concepts were considered to be “supernatural” or beyond understanding. Conclusion This study showed that traditional healers used conventional concepts and conventional practitioners could use traditional concepts. This suggests that conventional allopathic medical providers have the capacity to integrate biomedical concepts and other therapeutic and explanatory models. This information may be helpful to understand and improve risk management by anticipating and preventing potential reasons for failure in TM integration strategies and to enhance communication between patients, healers, and physicians to optimize TM or complementary and alternative medicine integration.

Towards an Electronic Dictionary of Tamajaq Language in Niger

We present the Tamajaq language and the dictionary we used as main linguistic resource in the two first parts. The third part details the complex morphology of this language. In the part 4 we describe the conversion of the dictionary into electronic form, the inflectional rules we wrote and their implementation in the Nooj software. Finally we present a plan for our future work.

Automatic natural acquisition of a semantic network for information retrieval systems

The amount of information is becoming greater and greater, in industries where complex processes are performed it is becoming increasingly difficult to profit from all the documents produced when fresh knowledge becomes available (reports, experiments, findings). This situation causes a considerable and expensive waste of precious time lost searching for documents or, quite simply, results in outright repeating what has been done. One solution is to transform all paper information into computerized information. We might imagine that we are in a science-fiction world and that we have the perfect computer. We tell it everything we know, we make it read all the books, and if we ask it any question, it will find the response if that response exists. But unfortunately, we are in the real world and the last four decades have taught us to minimize our expectations of computers. During the 1960s, the information retrieval systems appeared. Their purpose is to provide access to any desired documents, in response to a question about a subject, even if it is not known to exist. Here we focus on the problem of selecting items to index the documents. In 1966, Salton identified this problem as crucial when he saw that his system, Medlars, did not find a relevant text because of the wrong indexation. Faced with this problem, he imagined a guide to help authors choose the correct indexation, but he anticipated the automation of this operation with the SMART system. It was stated previously that a manual language analysis for information items by subjects experts is likely to prove impractical in the long run. After a brief survey of the existing responses to the index choice problem, we shall present the system automatic natural acquisition (ANA) which chooses items to index texts by using as little knowledge as possible- -just by learning the language. This system does not use any grammar or lexicon, so the selected indexes will be very close to the field concerned in the texts.