Abstract
A modified version of incremental learning model of Lebowitz's UNIMEM is proposed in this paper1. This new model is called G-UNIMEM which is motivated from the natural language acquisition study. It can extract causal relations from a set of training instances annotated with specified cause and goal features. The experiment has been carried out on the acquisition of anaphoric regularity in Mandarin Chinese.
Original language | English |
---|---|
Title of host publication | Proceedings - ICCI 1992 |
Subtitle of host publication | 4th International Conference on Computing and Information |
Editors | Peter E. Lauer, Anestis A. Toptsis, Waldemar W. Koczkodaj |
Publisher | Institute of Electrical and Electronics Engineers Inc. |
Pages | 278-281 |
Number of pages | 4 |
ISBN (Electronic) | 081862812X, 9780818628122 |
DOIs | |
State | Published - 1992 |
Externally published | Yes |
Event | 4th International Conference on Computing and Information, ICCI 1992 - Toronto, Canada Duration: 28 05 1992 → 30 05 1992 |
Publication series
Name | Proceedings - ICCI 1992: 4th International Conference on Computing and Information |
---|
Conference
Conference | 4th International Conference on Computing and Information, ICCI 1992 |
---|---|
Country/Territory | Canada |
City | Toronto |
Period | 28/05/92 → 30/05/92 |
Bibliographical note
Publisher Copyright:©1992 IEEE.
Keywords
- Anaphora
- Case-based learning
- Incremental learning
- Natural language acquisition