Ideas for Google Summer of Code

From Apertium
Revision as of 20:55, 12 February 2015 by Firespeaker (talk | contribs) (→‎List)
Jump to navigation Jump to search

Contents

This is the ideas page for Google Summer of Code, here you can find ideas on interesting projects that would make Apertium more useful for people and improve or expand our functionality. If you have an idea please add it below, if you think you could mentor someone in a particular area, add your name to "Interested mentors" using ~~~

The page is intended as an overview of the kind of projects we have in mind. If one of them particularly piques your interest, please come and discuss with us on #apertium on irc.freenode.net, mail the mailing list, or draw attention to yourself in some other way.

Note that, if you have an idea that isn't mentioned here, we would be very interested to hear about it.

Here are some more things you could look at:

List

Bring a released language pair up to state-of-the-art quality
How ?
(required skills)
What ?
(description)
Why ?
(rationale)
Who ?
(mentors)
XML, a scripting language (Python, Perl), good knowledge of the language pair adopted. Take a released language pair, and drastically improve the performance both in terms of coverage, and in terms of translation quality. This will involve working with dictionaries, transfer rules, scripting, corpora. The objective is to make an Apertium language pair state-of-the-art, or close to state-of-the-art in terms of translation quality. This will involve improving coverage to 95-98% on a range of corpora and decreasing word error rate by 30-50%. For example if the current word error rate is 30%, then it should be reduced to 15-20%. Apertium has quite a broad coverage of language pairs, but few of these pairs offer state-of-the-art translation quality. We think broad is important, but deep coverage is important too. Francis Tyers
2. Medium read more...
Adopt an unreleased language pair
How ?
(required skills)
What ?
(description)
Why ?
(rationale)
Who ?
(mentors)
XML, a scripting language (Python, Perl), good knowledge of the language pair adopted. Take on an orphaned unreleased language pair, and bring it up to release quality results. What this quality will be will depend on the language pair adopted, and will need to be discussed with the prospective mentor. This will involve writing linguistic data (including morphological rules and transfer rules — which are specified in a declarative language — and possibly Constraint Grammar rules if that is relevant) Apertium has a few pairs of languages (e.g. mt-he, ga-gd, ur-hi, pl-cs, sh-ru, etc...) that are orphaned, they don't have active maintainers. A lot of these pairs have a lot of work already put in, just need another few months to get them to release quality. See also Incubator Francis Tyers, Jimregan, Kevin Scannell, Trondtr, Unhammer, Darthxaher, Firespeaker, Hectoralos, Hrvoje Peradin, Jacob Nordfalk
3. Entry level read more...
Extend lttoolbox to have the power of HFST
How ?
(required skills)
What ?
(description)
Why ?
(rationale)
Who ?
(mentors)
C++, XSLT, XML Extend lttoolbox (perhaps writing a preprocessor for it) so that it can be used to do the morphological transformations currently done with HFST. And yes, of course, writing something that translates the current HFST format to the new lttolbox format. Proof of concept: Come up with a new format that can express all of the features found in the Kazakh transducer; implement this format in Apertium; Implement the Kazakh transducer in this format and integrate it in the English--Kazakh pair. Some language pairs in Apertium use HFST where most language pairs use Apertium's own lttoolbox. This is due to the fact that writing morphologies for languages that have features such as the vowel harmony found in Turkic languages is very hard with the current format supported by lttoolbox. The mixture of HFST and lttoolbox makes it harder for people to develop some language pairs. Mikel Forcada, Tommi A Pirinen, mentors wanted!
1. Hard read more
Discontiguous multiwords
How ?
(required skills)
What ?
(description)
Why ?
(rationale)
Who ?
(mentors)
C++, Knowledge of FSTs The task will be to develop, or adapt a module to deal with these kind of contiguous multiword expressions, for example, taking 'liggja ekki fyrir' and reordering it as 'liggja# fyrir ekki'. In many languages, such as English, Norwegian and Icelandic, there are discontiguous multiwords, e.g. phrasal verbs, that we cannot easily support. For example 'liggja ekki fyrir' in Icelandic should be translated in English as 'to be not clear', but we cannot have 'liggja fyrir' as a traditional multiword because of the extra 'adverb', or it could even be a whole NP. Francis Tyers
2. Medium read more...
Flag diacritics in lttoolbox
How ?
(required skills)
What ?
(description)
Why ?
(rationale)
Who ?
(mentors)
C++ or Java, XML, Knowledge of FSTs Adapt lttoolbox to elegantly use flag diacritics. Flag diacritics are a way of avoiding transducer size blow-up by discarding impossible paths at runtime as opposed to compile time. Some work has already been done, see Flag diacritics. This will involve designing some changes to our XML dictionary format (see lttoolbox, and implementing the associated changes in the FST compiling processing code. The reason behind this is that many languages have prefix inflection, and we cannot currently deal with this without either making paradigms useless, or overanalysing (e.g. returning analyses where none exist). Flag diacritics (or constraints) would allow us to restrict overanalysis without blowing up the size of our dictionaries. Francis Tyers (C++), Jacob Nordfalk (Java)
1. Hard read more...
lint for Apertium
How ?
(required skills)
What ?
(description)
Why ?
(rationale)
Who ?
(mentors)
Python, C++, XML, autotools Make a program which tests Apertium data files for suspicious or unrecommended constructs (likely to be bugs). Somtimes when several people are working on the same code, things can get repeated, or beginners can make unrecommended changes. A lint tester would help people write standard code for dictionaries and transfer files. Francis Tyers
3. Entry level read more...
Robust recursive transfer
How ?
(required skills)
What ?
(description)
Why ?
(rationale)
Who ?
(mentors)
Python, XML, linguistics The purpose of this task would be to create a module to replace the apertium-transfer module(s) which will parse and allow transfer operations on an input. Currently we have a problem with very distantly related languages that have long-distance constituent reordering, because we can only do finite-state chunking. Francis Tyers, Sortiz
1. Hard read more...
Weighted transfer rules
How ?
(required skills)
What ?
(description)
Why ?
(rationale)
Who ?
(mentors)
Python, C++, linguistics The purpose of this task is to add weights to our transfer rules and allow conflicting rule patterns to be handled by combining (lexicalised) weights. Currently our transfer rules are applied longest-match left-to-right (LRLM). When two rule patterns conflict the first one is chosen. Francis Tyers, TommiPirinen
1. Hard read more...
Extend Mitzuli to support all language pairs
How ?
(required skills)
What ?
(description)
Why ?
(rationale)
Who ?
(mentors)
C++, Java, Android Integrate runtimes for HFST and vislcg3 into Mitzuli, an Android app for translation. Mitzuli (see here) is a fantastic Android app for mobile telephones. It is an interface, based on lttoolbox-java and other programs that allows you to translate, and do OCR combined with translation. The main drawback at the moment is that it doesn't support HFST and also doesn't support CG, which a number of language pairs rely on (for example Sámi, Turkic, Breton, Welsh etc.) The objective of this task is to make Mitzuli support them. Mikel, Tino Didriksen, TommiPirinen
1. Hard read more...
Automatic blank handling
How ?
(required skills)
What ?
(description)
Why ?
(rationale)
Who ?
(mentors)
C++, XML Let the C++ modules deal with formatting, treating some tags as word-bound Our current format handling is brittle, requiring transfer rules to explicitly deal with blanks, and some times inevitably outputting them in the wrong order. This project is to do blank-handling automatically, by treating some blanks as "glued to words" (e.g. italics, emphasis), and others as paragraph-level blanks. Unhammer, Tino Didriksen
2. Medium read more...
Integration and debugging tools for Grammatical Framework
How ?
(required skills)
What ?
(description)
Why ?
(rationale)
Who ?
(mentors)
Haskell, C, XML The objective of this task is to create a suite of tools for Grammatical Framework (GF) to facilitate interoperability with Apertium and other tools. Grammatical Framework has a ton of resources, but people familiar with Apertium may find the way of using them a bit alien. The idea of this task is to make some tools for GF to make it more easily pick-uppable for Apertium developers and users. Francis Tyers
2. Medium read more...
Weights in lttoolbox
How ?
(required skills)
What ?
(description)
Why ?
(rationale)
Who ?
(mentors)
C++, XML, FSTs lttoolbox is a set of tools for building finite-state transducers. It would be good to add possibility for weighting lexemes and analyses. Weighting information for lexical forms could be useful for the part-of-speech tagger. Francis Tyers, TommiPirinen
2. Medium read more...
Improvements to the Apertium website
How ?
(required skills)
What ?
(description)
Why ?
(rationale)
Who ?
(mentors)
Python, HTML, JS ... ... Francis Tyers
3. Entry level read more...
Develop a spell checking system
How ?
(required skills)
What ?
(description)
Why ?
(rationale)
Who ?
(mentors)
Python, HTML, JS Apertium hfst transducers can be compiled into libraries that libvoikko can use to perform spell checking, including providing suggestions. Additionally, we have the beginnings of a spell checking interface on our website. This project would consist of (1) determining the best way to make spell-checking integrated into apertium, both as a configurable option to compile, and as an apertium mode (integrate this into apertium-init); (2) making APY support this type of mode; and (3) finishing the web interface for html-tools for spell checking and making it interface with APY.
One extra, but more difficult, sub-project that would be useful is to make sure that both hfst and lttoolbox transducers compile to libvoikko, or at least converting between lttoolbox and hfst formats so that both compile to libvoikko format (currently hfst transducers can be compiled to lttoolbox transducers)
Spell checking is one of more useful services our language resources can provide. Being able to easily provide spell checking is important both for language communities to be able to use our resources and as a new entry point for prospective developers. Jonathan
2. Medium read more...