Reading: Multi-Prototype Vector-Space Models of Word Meaning

[a1da4]

0. Paper

@inproceedings{reisinger-mooney-2010-multi, title = "Multi-Prototype Vector-Space Models of Word Meaning", author = "Reisinger, Joseph and Mooney, Raymond J.", booktitle = "Human Language Technologies: The 2010 Annual Conference of the North {A}merican Chapter of the Association for Computational Linguistics", month = jun, year = "2010", address = "Los Angeles, California", publisher = "Association for Computational Linguistics", url = "https://aclanthology.org/N10-1013", pages = "109--117", }

1. What is it?

2. What is amazing compared to previous works?

3. Where is the key to technologies and techniques?

They model contextualized information in each occurrence using features of around 10 words.

• chi^2
• tf-idf

They define two similarity functions.

• normal

• contextual information is availabled

4. How did evaluate it?

Figure 2 shows that:

• Feature: tf-idf achieved higher performance than chi^2
• Function: Avg achieved higher performance than Max
• Prototype: multi-prototype (this paper) achieved high performance than a single prototype

From Table 1, multi-prototype achieved higher than single prototype (K=1) and exemplar (K=C, frequency of words).

5. Is there a discussion?

6. Which paper should read next?

[a1da4]

202

Using neural language model to consider local and global information (the multi-prototype method is based on this paper)