• No results found

Using WordNet and Semantic Similarity for Bilingual Terminology Mining from Comparable Corpora

N/A
N/A
Info
Download
Protected

Academic year: 2020

Share "Using WordNet and Semantic Similarity for Bilingual Terminology Mining from Comparable Corpora"

Copied!
8
0
0

Loading.... (view fulltext now)

Download now ( 8 Page )

Full text

Loading

Figure

Figure 1: Overall architecture of the lexical extraction approach
Table 2: Comparable corpora sizes in term ofwords.
Figure 2: Wikipedia categories tree of the corporate finance sub-domain.
Table 3: Precision, Recall at TopN(N=1,10,20) and MRR at Top20 for the two domains. In each column,bold show best results
+2

References

Download now ( PDF - 8 Page - 354.67 KB )

Related documents

Using a Bilingual Resource to Add Synonyms to a Wordnet : FinnWordNet and Wikipedia as an Example

3 Method for Finding Synonym Candidates in a Bilingual Resource Our method essentially mines new synonyms for a wordnet by using translation pairs in a bilingual re- source

Improving MT System Using Extracted Parallel Fragments of Text from Comparable Corpora

While searching for parallel chunks each translated source chunk was compared with all the target chunks in the corresponding document.. Table 2 displays the number

Combining String and Context Similarity for Bilingual Term Alignment from Comparable Corpora

(2014) introduced a Random Forest (RF) classifier that learns correspondences of character n-grams between a source and target language. Unlike lexical and sub-lexical compo-