scholarly journals Lexical Knowledge Acquisition Using Spontaneous Descriptions in Texts

Author(s):  
Augusta Mela ◽  
Mathieu Roche ◽  
Mohamed El Amine Bekhtaoui
Author(s):  
Alfio Massimiliano Gliozzo ◽  
Aditya Kalyanpur

Automatic open-domain Question Answering has been a long standing research challenge in the AI community. IBM Research undertook this challenge with the design of the DeepQA architecture and the implementation of Watson. This paper addresses a specific subtask of Deep QA, consisting of predicting the Lexical Answer Type (LAT) of a question. Our approach is completely unsupervised and is based on PRISMATIC, a large-scale lexical knowledge base automatically extracted from a Web corpus. Experiments on the Jeopardy! data shows that it is possible to correctly predict the LAT in a substantial number of questions. This approach can be used for general purpose knowledge acquisition tasks such as frame induction from text.


1989 ◽  
Vol 4 (1) ◽  
pp. 1-20 ◽  
Author(s):  
F. Antonaci ◽  
M. Russo ◽  
M.T. Pazienza ◽  
P. Velardi

2014 ◽  
Vol 9 ◽  
Author(s):  
Lenhart Schubert

Recent implementations of Natural Logic (NLog) have shown that NLog provides a quite direct means of going from sentences in ordinary language to many of the obvious entailments of those sentences. We show here that Episodic Logic (EL) and its Epilog implementation are well-adapted to capturing NLog-like inferences, but beyond that, also support inferences that require a combination of lexical knowledge and world knowledge. However, broad language understanding and commonsense reasoning are still thwarted by the “knowledge acquisition bottleneck”, and we summarize some of our ongoing and contemplated attacks on that persistent difficulty.


1996 ◽  
Vol 9 (2) ◽  
pp. 79-82 ◽  
Author(s):  
Marc Domenig ◽  
Alain Hsiung

1992 ◽  
Author(s):  
Takehito Utsuro ◽  
Yuji Matsumoto ◽  
Makoto Nagao

1990 ◽  
Vol 55 (4) ◽  
pp. 665-678 ◽  
Author(s):  
Beth A. Mineo ◽  
Howard Goldstein

This study examined the effectiveness of matrix-training procedures in teaching action + object utterances in both the receptive and expressive language modalities. The subjects were 4 developmentally delayed preschool boys who failed to produce spontaneous, functional two-word utterances. A multiple baseline design across responses with a multiple probe technique was employed. Subjects were taught 4–6 of 48 receptive and 48 expressive responses. Acquisition of a word combination rule was facilitated by the use of familiar lexical items, whereas subsequent acquisition of new lexical knowledge was enhanced by couching training in a previously trained word combination pattern. Although receptive knowledge was not sufficient for the demonstration of corresponding expressive performance for most of the children, only minimal expressive training was required to achieve this objective. For most matrix items, subjects responded receptively before they did so expressively. For 2 subjects, when complete receptive recombinative generalization had not been achieved, expressive training facilitated receptive responding. The results of this study elucidate benefits to training one linguistic aspect (lexical item, word combination pattern) at a time to maximize generalization in developmentally delayed preschoolers.


Sign in / Sign up

Export Citation Format

Share Document