The IUCL+ System: Word-Level Language Identification via Extended Markov Models

@inproceedings{King2014TheIS,
  title={The IUCL+ System: Word-Level Language Identification via Extended Markov Models},
  author={Levi King and Eric Baucom and T. Gilmanov and Sandra K{\"u}bler and Daniel Whyatt and Wolfgang Maier and Paul Rodrigues},
  booktitle={CodeSwitch@EMNLP},
  year={2014}
}
We describe the IUCL+ system for the shared task of the First Workshop on Computational Approaches to Code Switching (Solorio et al., 2014), in which participants were challenged to label each word in Twitter texts as a named entity or one of two candidate languages. Our system combines character n-gram probabilities, lexical probabilities, word label transition probabilities and existing named entity recognitiontools within a Markovmodel framework that weights these components and assigns a… Expand
16 Citations
LILI: A Simple Language Independent Approach for Language Identification
  • 15
  • PDF
Overview for the Second Shared Task on Language Identification in Code-Switched Data
  • 88
  • Highly Influenced
  • PDF
Overview for the First Shared Task on Language Identification in Code-Switched Data
  • 170
  • Highly Influenced
  • PDF
AIDA2: A Hybrid Approach for Token and Sentence Level Dialect Identification in Arabic
  • 23
  • PDF
A deep learning approach for the romanized tunisian dialect identification
  • PDF
Recurrent-Neural-Network for Language Detection on Twitter Code-Switching Corpus
  • 16
  • PDF
Code-Mixing: A Brief Survey
  • S. Thara, P. Poornachandran
  • Computer Science
  • 2018 International Conference on Advances in Computing, Communications and Informatics (ICACCI)
  • 2018
  • 9
Segregation of Code-Switching Sentences using Rule-Based Technique
...
1
2
...

References

SHOWING 1-10 OF 13 REFERENCES
Word-level language identification in The Chymistry of Isaac Newton
  • 9
  • PDF
Arabic Named Entity Recognition using Conditional Random Fields
  • 103
Language Identifier: A Computer Program for Automatic Natural-Language Identification of On-line Tex
  • 121
  • Highly Influential
Incorporating Non-local Information into Information Extraction Systems by Gibbs Sampling
  • 3,023
  • PDF
...
1
2
...