• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 1
  • Tagged with
  • 2
  • 2
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
1

Mise en œuvre des préférences dans des problèmes de décision / Preference Handling in Decision-Making Problems

Patel, Namrata 07 October 2016 (has links)
Il y a une forte croissance, à nos jours, de «services» intelligents proposés aux clients sur les plates-formes de commerce électronique, destinés à une assistance personnalisée. L'étude de préférences a suscité un grand intérêt dans ce contexte, grâce à leur utilisation dans la résolution de problèmes liés à la prise de décision. En effet, la recherche sur les préférences en intelligence artificielle (IA) propose différentes manières d'aborder ce problème : de l'acquisition des préférences à leur représentation formelle et, éventuellement, à leur gestion suivant plusieurs méthodes de raisonnement. Dans cette thèse, nous adressons la problématique de la mise en œuvre de préférences comparatives pour l'aide à la décision par le développement d'un système interactif «intelligent» de recommandations personnalisées. Nous suivons une tendance récente, et le concevons sur une base de considérations psychologiques, linguistiques et personnelles. Nous contribuons ainsi aux domaines suivants de préférences en IA : (1) leur acquisition, (2) leur représentation, et (3) leur mise en œuvre. Nous examinons d'abord un goulot d'étranglement dans l'acquisition de préférences et proposons une méthode d'acquisition de préférences exprimées en langage naturel (LN), qui permet leur représentation formelle en tant que préférences comparatives. Nous étudions ensuite les aspects théoriques de la représentation et du raisonnement avec les préférences comparatives pour aide à la décision. Finalement, nous décrivons notre outil de recommandations qui utilise : (1) une base de données de produits qualifiée par une analyse de critiques d'utilisateurs, (2) une approche interactive pour guider les utilisateurs à exprimer leurs préférences comparatives, et (3) un moteur de raisonnement qui manipule ces préférences afin de proposer une recommandation basée sur les préférences de l'utilisateur. / Intelligent ‘services’ are increasingly used on e-commerce platforms to provide assistance to customers. In this context, preferences have gained rapid interest for their utility in solving problems related with decision making. Research on preferences in AI has shed light on various ways of tackling this problem, ranging from the acquisition of preferences to their formal representation and eventually their proper manipulation. Following a recent trend of stepping back and looking at decision-support systems from the user’s point of view, i.e. designing them on the basis of psychological, linguistic and personal considerations, we take up the task of developing an “intelligent” tool which uses comparative preference statements for personalised decision support. We tackle and contribute to different branches of research on preferences in AI: (1) their acquisition (2) their formal representation and manipulation (3) their implementation. We first address a bottleneck in preference acquisition by proposing a method of acquiring user preferences, expressed in natural language (NL), which favours their formal representation and further manipulation. We then focus on the theoretical aspects of handling comparative preference statements for decision support. We finally describe our tool for product recommendation that uses: (1) a review-based analysis to generate a product database, (2) an interactive preference elicitation unit to guide users to express their preferences, and (3) a reasoning engine that manipulates comparative preference statements to generate a preference-based ordering on outcomes as recommendations.
2

A shallow processing approach to anaphor resolution

Carter, David Maclean January 1986 (has links)
The thesis describes an investigation of the feasibility of resolving anaphors in natural language texts by means of a "shallow processing" approach which exploits knowledge of syntax, semantics and local focussing as heavily as possible; it does not rely on the presence of large amounts of world or domain knowledge, which are notoriously hard to process accurately. The ideas reported are implemented in a program called SPAR (Shallow Processing Anaphor Resolver), which resolves anaphoric and other linguistic ambiguities in simple English stories and generates sentence-by-sentence paraphrases that show what interpretations have been selected. Input to SPAR takes the form of semantic structures for single sentences constructed by Boguraev's English analyser. These structures are integrated into a network-style text representation as processing proceeds. To achieve anaphor resolution, SPAR combines and develops several existing techniques, most notably Sidner's theory of local focussing and Wilks' "preference semantics" theory of semantics and common sense inference. Consideration of the need to resolve several anaphors in the same sentence results in Sidner's framework being modified and extended to allow focus-based processing to interact more flexibly with processing based on other types of knowledge. Wilks' treatment of common sense inference is extended to incorporate a wider range of types of inference without jeopardizing its uniformity and simplicity. Further, his primitive-based formalism for word sense meanings is developed in the interests of economy, accuracy and ease of use. Although SPAR is geared mainly towards resolving anaphors, the design of the system allows many non-anaphoric (lexical and structural) ambiguities that cannot be resolved during sentence analysis to be resolved as a by-product of anaphor resolution.

Page generated in 0.0524 seconds