HowNet and the computation of meaning

Zhendong Dong

Linguistic Knowledge Dept,Computer & Language Information Research Centre Chinese Academy of Sciences , China

Le 22 novembre 2006 à 11 h 30

Salle 3195, Pavillon André-Aisenstadt

We will give a panoramic picture of HowNet as a knowledge system, presenting its data structure of Chinese/English words and expressions and its hierarchical taxonomies of sememes. Through the HowNet's browser, we demonstrate various kinds of relations among concepts computed on the basis of concept definitions. Two HowNet- based tools, Concept Similarity Measure (CSM) and Concept Relevance Computation (CRC) show the usefulness of sememes in meaning computation. We review the extraction of sememes in the development of HowNet and discuss the success in the scheme of sememes (or primitives).

Further details can be found in the speakeru2019s book "HowNet and the Computation of meaning" published in April 2006.

Pour recevoir les annonces hebdomadaires par courriel, visitez

Liste des autres séminaires