Resources


Data

Internal Software
  • Perl scripts for training GMTK DBN models for transliteration scoring (which I modified from Filali Bilmes' scripts (2006)). Please contact me for these.
  • C++ implementations for Pair Hidden Markov Models and Perl scripts for transliteration scoring (which I modified from Mackay and Kondrak (2004), & Wieling et al.'s (2007) implementations). Please contact me for these.

External software that I use (or have used) in my research/projects/assignments
  • Graphical models: The Graphical Models Toolkit (GMTK) - I used GMTK in my PhD research for modeling transduction-based dynamic Bayesian networks that i applied in the NLP sub task of transliteration detection.
  • Statistical Machine Translation: MOSES phrase-based MT system, GIZA++ (for word alignment), IRSTLM & SRILM & KenLM (for natural language modeling) - I used all these in my PhD research in the NLP sub task of transliteration generation. I also use them now in Machine Translation research.
  • Neural Machine Translation: OpenNMT - I use this in Machine Translation research.
  • Finite state automata: Carmel finite state toolkit - I used Carmel in my PhD research in the NLP sub task of transliteration generation.
  • Expert Systems: Dezide Advisor - I used an evaluation copy (unfortunately it's commercial) of this tool in my MSc Computer Science project to build Bayesian network models for troubleshooting heavy duty diesel engine faults.
  • MeTA (ModErn Text Analysis) - a modern C++ Data Sciences Toolkit that I used in several tasks of the Data mining specialization course from University of Illinois at Urbana Champaign.
  • R: for statistical computing and graphics - I often use this for plotting graphs.

External software that is being (or has been) used under my supervision
  • Expert Systems: expertise2go - for implementing rule-based expert systems - has Android version as well for implementing mobile apps of expert systems.
  • Weka classification software

Subpages (2): NLP data Parallel Corpora
Comments