LABDA-UC3M

 

  • Paloma Martínez (PI)
  • Harith Al-Jumaily
  • Fco. Javier Calle
  • Elena Castro
  • Dolores Cuadra
  • Ana María Iglesias
  • Leonardo Castaño
  • Lourdes Moreno
  • Julián Moreno
  • Isabel Segura
  • María González
  • Daniel Sánchez
  • María Herrero
  • Rocío Calvo

 

Laboratorio de Bases de Datos Avanzadas at Universidad Carlos III de Madrid has participated in recent years in projects related to natural language processing, natural interaction, information retrieval, answers search systems and storage technologies.

Among the latest projects related to language technologies, we can mention: MIRACLE (CM 07T/0055/2003 2), consisting on the development of a Multilingual Information Retrieval System and its assessment by the CLEF forum; GPS Platform for Software Process Management: Modeling, reutilisation and measuring (TIN2004-07083) in which language processing techniques were applied to documents created in software projects; BRAVO: Multimodal and Multilingual Advanced Answers Search (TIN2007-67407-C03-01) devoted to research on technologies to improve the answers search in both text and voice.

LABDA-UC3M participates in the MAVIR-CM programme as well as in the Thematic Network on Multilingual and Multimodal Information Processing (TIMM, TSI2007-30967-E). In addition, some researchers are part of the Centro Español de Subtitulado y Audiodescripción (CESYA) under the Royal
Patronage on Disability to promote initiatives that enable people with visual and hearing disabilities to access television contents. In CESYA, the group participated in the definition of IRIS, a semi-automatic consultation, validation and integration service of information on subtitled and/or audiowritten audiovisual materials.

Furthermore, together with the Hospital de Fuenlabrada, the group participated in the project  ISSE: Semantic-based Interoperability for eHealth (FIT-350300-2007-75), in which the aim is to research in technology based on automatic language processing for information location and retrieval from medical texts and other resources (reports, electronic medical records, scientific documentation, etc..) specially in Spanish. Thus, we worked in browsers with different levels of complexity that integrate medical domain-specific resources and terminology (UMLS, SNOMED, etc.) and with different treatment of syntactic and semantic levels.One of the current research areas is related to the location of drug names and interactions between them in the medical literature using UMLS, dictionaries and USAN rules of naming drugs. As a result, the group has generated an annotated corpus with generic drug names and other biomedical concepts. This corpus, which has been manually revised by a pharmacological expert, consists of 849 abstracts downloaded from PubMed.

In the line of natural interaction, the group participated in the following projects: INTEGRA TV-4ALL. Interactive Television for the Blinds"  (FIT-350301-2004-2), in which a natural language dialogue manager has been developed for people with visual disabilities in the environment of a hotel room in order to access certain services; SOPAT (CIT-410000-2007-12) "Personalised and Accessible Guidance Service for Tourism”, aimed at improving the existing signal systems for tourists of all languages and developing, in tourist establishments, help systems for disabilities (motor, auditory and visual). Access Channel to Resources and Digital Contents (IST-020501-2008-54) aims at developing a new digital channel for complete information and services access in business areas; Thuban: Plataforma de Interacción Natural para el acompañamiento virtual en entornos reales (TIN2008-02711) aims at the implementation of a virtual companion capable of interacting with the user while moving in an environment (with a state model based in spatio-temporal databases for managing positioning), providing access to a predetermined set of services that take into account the user’s position and
trajectory and oriented to manage these parameters and any service available on the system (e.g. searching in a catalogue).

As a result of research in these lines, LABDA-UC3M has generated two tools which are part of the of the catalogue Oferta Tecnológica, Servicios Técnicos y Programas de I+D published in 2008 by madri+d. INTERACTOR is a natural interaction platform including a dialogue system and a situation
component based on spatio-temporal databases. On the other hand "SQUASH: is a Question Answering System for Spanish”, jointly developed by researchers from the group and UAM Laboratorio de Lingüística Informática. SQUASH is able to provide precise answers in real time to questions in natural language such as When was the Maastricht treaty signed? The system is designed in modules: it includes a set of rules to select the type of information likely to be relevant from a given question, question analysis techniques, information extraction and information retrieval.

 

 
RocketTheme Joomla Templates