[ Home - Publications - Softwares - Teaching ]

Softwares

ALANIS: A Learning by ANalogy Inferencer for Structured data
  • ALANIS implements a supervised learning algorithm based on analogical proportions capable of dealing with structured data such as strings, trees, and feature structures.
    Its main area of application is Natural Language Learning. For example, it can be used to infer the morphological analysis (seen as a tree or a feature structure) or the pronunciation (seen as a string) of a given wordform.
    It needs a training file of pairs of (input,output), and an input file of inputs to analyze. For a given input, possibly several (or no) solutions are proposed.
    See the following references for more information:
    - Nicolas Stroppa and François Yvon. An analogical learner for morphological analysis. In Proceedings of the 9th Conference on Computational Natural Language Learning (CoNLL 2005), pages 120–127, Ann Arbor, MI, June 2005. Association for Computational Linguistics.
    - Nicolas Stroppa. Définitions et caractérisations de modèles à base d'analogies pour l'apprentissage automatique des langues naturelles. PhD Thesis. École Nationale Supérieure des Télécommunications, Paris, Novembre 2005.

    This sofware is released under a GNU General Public License.
  • README.
  • Last version: alanis-0.9.7.
FastMtEval
  • FastMtEval implements several automatic Machine Translation evaluation metrics: BLEU, NIST, and GTM-1. It supports bootstrap evaluation of confidence intervals, as well as statistical significance testing of differences between two systems. It is written in C++ and should be very fast.

    This sofware is released under a GNU General Public License.
  • README.
  • Last version: fast_mt_eval.
STE: a Simple Test Environment
  • STE is a simple test environment whose goal is to make easier the task of performing experiments with different sets of parameters. the idea of STE is _very_ simple: it creates one directory when you need one and associates one unique id to it. When you create such a directory, you have to label it with a list of keywords. Directories can then be addressed by those keywords.
    This sofware is released under a GNU General Public License.
  • README.
  • Last version: ste-1.0.
ANHI: A Naive HMM Implementation
  • Anhi is a naive HMM implementation whose primary goal is to align sequences of graphemes and sequences of phonemes using an Hidden Markov Model and the EM algorithm.
    This sofware is released under a GNU General Public License.
  • README.
  • Last version: anhi-0.2.
Celex_to_Sqlite
  • celex_to_sqlite is a tool to convert the Celex lexicons into an sqlite database. It makes easier and faster the execution of queries.
    This sofware is released under a GNU General Public License.
  • README.
  • Last version: celex_to_sqlite-0.2.
[ Home - Publications - Softwares - Teaching ]