Name
Abstract | ||
|---|---|---|
This paper presents a new distributed representation based approach to capture commonsense knowledge. Currently, we focus on meanings of verbs and related dependency relations. Built upon such dependency relations, a set of vector space models are developed to capture the meaning of the verb and transfer it to related ambiguous pronouns.To demonstrate the capability of this data-driven approach, we do not incorporate any human knowledge. Among all the vector models without human knowledge, we can achieve the best performance on all of the verb similarity datasets. We can also achieve superior performance on a subset of Winograd Schema challenge dataset, compared with other existing pronoun co-reference and embedding models. |
Year | DOI | Venue |
|---|---|---|
2018 | 10.1145/3195106.3195127 | PROCEEDINGS OF 2018 10TH INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND COMPUTING (ICMLC 2018) |
Keywords | Field | DocType |
Commonsense Knowledge, Word Embedding, Pronoun Co-reference Resolution, Winograd Schema Challenge | Pronoun,Verb,Commonsense knowledge,Vector space,Embedding,Winograd Schema Challenge,Computer science,Natural language processing,Human knowledge,Artificial intelligence,Word embedding,Machine learning | Conference |
Citations | PageRank | References |
0 | 0.34 | 15 |
Authors | ||
2 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Hongming Zhang | 1 | 10 | 8.34 |
| Yangqiu Song | 2 | 2045 | 103.29 |