1999
DOI: 10.1023/a:1007537716579
|View full text |Cite
|
Sign up to set email alerts
|

Untitled

Abstract: Abstract. In many applications of natural language processing (NLP) it is necessary to determine the likelihood of a given word combination. For example, a speech recognizer may need to determine which of the two word combinations "eat a peach" and "eat a beach" is more likely. Statistical NLP methods determine the likelihood of a word combination from its frequency in a training corpus. However, the nature of language is such that many word combinations are infrequent and do not occur in any given corpus. In … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
12
0

Year Published

2005
2005
2022
2022

Publication Types

Select...
4
3
3

Relationship

0
10

Authors

Journals

citations
Cited by 235 publications
(12 citation statements)
references
References 46 publications
0
12
0
Order By: Relevance
“…It is clear that the effectiveness of ⊔ int increases as the size of C grows, and that it would particularly benefit from distributional smoothing (Dagan, Pereira, and Lee, 1994) which can be used to improve plausible co-occurrence coverage by inferring co-occurrences in the Apt for a lexeme w based on the co-occurrences in the Apts of distributionally similar lexemes.…”
Section: Merging Aligned Aptsmentioning
confidence: 99%
“…It is clear that the effectiveness of ⊔ int increases as the size of C grows, and that it would particularly benefit from distributional smoothing (Dagan, Pereira, and Lee, 1994) which can be used to improve plausible co-occurrence coverage by inferring co-occurrences in the Apt for a lexeme w based on the co-occurrences in the Apts of distributionally similar lexemes.…”
Section: Merging Aligned Aptsmentioning
confidence: 99%
“…Improved from VSM, a better word representation method named latent semantic analysis(LSA) was proposed [24], it assumed that words that are close in meaning will occur in similar pieces of text, and used global matrix factorization to generate word vector. Some other researchers also considered the word co-occurrence for better word representation [25][26][27]. The further approaches are to learn word representations within local context windows and neural network, e.g., a word representation method based on context windows and deep learning architectures was proposed in [28].…”
Section: Related Workmentioning
confidence: 99%
“…where 𝐾 is the number of topics and 𝑝, 𝑞 are the topics distributions It can be transformed into a similarity measure as follows [13] :…”
Section: Similarity Measures Across Documentsmentioning
confidence: 99%