Proceedings of the Annual Meeting of the Association for Computational Linguistics

Themes

Themes are sets of two or more topics that frequently appear together in articles. These are the themes found in articles published in Proc. ACL.

1. 93 papers (9%) mention
   1. machine translation
   2. BLEU
2. 82 papers (8%) mention
   1. language model
   2. BLEU
3. 47 papers (4%) mention
   1. Treebank
   2. POS tags
4. 42 papers (4%) mention
   1. bigram
   2. unigram
5. 42 papers (4%) mention
   1. sentence pairs
   2. word alignment
6. 41 papers (4%) mention
   1. n-gram
   2. n-grams
7. 40 papers (4%) mention
   1. dependency parsing
   2. dependency tree
8. 38 papers (3%) mention
   1. topic models
   2. LDA
9. 35 papers (3%) mention
   1. phrase-based
   2. translation model
10. 34 papers (3%) mention
    1. machine translation
    2. language model
    3. translation model
11. 31 papers (3%) mention
    1. labeled data
    2. unlabeled data
12. 30 papers (3%) mention
    1. WordNet
    2. synsets
13. 29 papers (2%) mention
    1. log-linear
    2. log-linear model
14. 27 papers (2%) mention
    1. part-of-speech
    2. part-of-speech tags
15. 27 papers (2%) mention
    1. role labeling
    2. semantic role
16. 24 papers (2%) mention
    1. coreference
    2. coreference resolution
17. 24 papers (2%) mention
    1. NER
    2. named entity
18. 22 papers (2%) mention
    1. n-gram
    2. BLEU
    3. BLEU score
19. 19 papers (1%) mention
    1. Chinese word
    2. word segmentation
20. 18 papers (1%) mention
    1. phrase pair
    2. word alignment