Adding transliteration code as a sub project to cogcomp-nlp

[mayhewsw]

This is old code written by Jeff Pasternack in C#, ported by me into Java, and finally being migrated out of gitlab into github.

[danyaljj]

How about a slight modification to the readme?

CogComp Transliteration

Transliteration is the conversion of a given name in the source language (from source script) to a name in the target language (target script), such that the target language name is:

1. phonemically equivalent to the source name

मममबई → Mumbai

2. conforms to the phonology of the target language

नरिदेरनद → ਨਰਰਰਦਰ (नरिदरिम )

3. matches the user intuition of the equivalent of the source language
   name in the target language, considering the culture and orthographic
   character usage in the target language

ആലപപഴ (aalappuzha)→ Alappuzha

About this package

This is a Java port of Jeff Pasternack's C# code from "Learning Better Transliterations."

To run, look at examples in TestTransliteration
or Runner.

Further reading

You can checkout more details on this work:

@inproceedings{PasternackRo09a,
    author = {J. Pasternack and D. Roth},
    title = {Learning Better Transliterations},
    booktitle = {CIKM},
    month = {11},
    year = {2009},
    url = "http://cogcomp.org/papers/PasternackRo09a.pdf",
    funding = {MIAS},
    projects = {TL},
    comment = {State-of-the-art transliteration with unbounded substring-to-substring productions and capable of both discovery and generation.},
}

[mayhewsw]

@danyaljj this failed, but I don't know why. I can't login to the TeamCity site.

[danyaljj]

We have tested this locally and seems like the failure is due to NER issues. With that in mind, merging this.