Hello Mr. Tyers, Thank you for the quick response. I know my idea was a little out there. The wiki, however, suggested that "if you have a 'way out there' idea, don't panic! We're still interested, but we'll try to find a subset of it which is achievable in the time scale available." I thought I give my idea a at least a shot.
Second Idea: I come from a statistical based understanding of computational linguistics. I think that "Detect hidden unknown words by using the probabilities of the HMM-based part-of-speech tagger in Apertium" is a good fit given my skill set and experience. Good work can come out of integrating work on unsupervised learning of morphology ( http://linguistica.uchicago.edu/linguistica.html ) with Apertium's part-of-speech tagger in order to detect missing entries in the morphological dictionary, using Linguistica's results to inform the more likely part-of-speech tag. Lingustica is open source so it can be integrated into the Apertium pipeline. My work over the summer would be reimplementing Linguistica and integrate it with Apertium's part-of-speech tagger. Does this sound like a good idea? If so, I can come back with a more detailed plan outlining how I would integrate Lingustica in order to find missing lexical forms in the Apertium's dictionaries. If this does not sound like a good idea, given my skill set and interests, can you suggest another project? Relevant Link: http://linguistica.uchicago.edu/linguistica.html Best Wishes, Rafael On Tue, Mar 22, 2011 at 1:39 PM, Francis Tyers <[email protected]> wrote: > > El dt 22 de 03 de 2011 a les 12:22 -0500, en/na Rafael Turner va > escriure: >> Dear members of Apertium, >> >> My name is Rafael and I am a second year student at the University of >> Chicago. I was wondering if I could please have some feedback on my >> project proposal. A link to my resume, project idea and timeline is >> http://home.uchicago.edu/~turnersr/gsoc/apertium/Steven_Rafael_Turner.pdf >> >> I am interested in machine translation because of its >> interdisciplinary nature. Linguistics is divided into several distinct >> areas: phonetics, phonology, morphology, syntax, semantics, >> pragmatics, etc. . Each of these areas has evolved throughout the >> years by developing its own ideas and techniques, and has reached by >> now a remarkable degree of specialization. Machine translation, >> however, requires that various connections between these areas and the >> creation of actual ‘bridges’ between different linguistic branches. >> Many times it seems that methods from all fields must employed to >> capture the nuance of language, and this interplay of different points >> of view in a same subject has played a fundamental role in >> illuminating the nature of language. Machine translation is a breeding >> ground for translating and comparing different linguistic theories, >> discovering analogies, and providing an effective means for >> transferring results and techniques between distinct fields. That is >> why I find machine translation so exciting! >> >> Please consider my prior experience in natural language processing, >> programming, and conceptualizing mathematically as proper >> qualifications. >> >> Take care, >> Rafael >> >> P.S. Please let me know if this is way out there. I have other ideas >> related to implementing morphology learning algorithms that may fit >> more naturally into the Apertium framework. > > Isn't this basically a character-based SMT system ? Remember that > Apertium is by-and-large an RBMT project -- have you looked over our > ideas on the ideas page?[1] > > Fran > > 1. http://wiki.apertium.org/wiki/Ideas_for_Google_Summer_of_Code > > > ------------------------------------------------------------------------------ > Enable your software for Intel(R) Active Management Technology to meet the > growing manageability and security demands of your customers. Businesses > are taking advantage of Intel(R) vPro (TM) technology - will your software > be a part of the solution? Download the Intel(R) Manageability Checker > today! http://p.sf.net/sfu/intel-dev2devmar > _______________________________________________ > Apertium-stuff mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/apertium-stuff > On Tue, Mar 22, 2011 at 1:39 PM, Francis Tyers <[email protected]> wrote: > > El dt 22 de 03 de 2011 a les 12:22 -0500, en/na Rafael Turner va > escriure: >> Dear members of Apertium, >> >> My name is Rafael and I am a second year student at the University of >> Chicago. I was wondering if I could please have some feedback on my >> project proposal. A link to my resume, project idea and timeline is >> http://home.uchicago.edu/~turnersr/gsoc/apertium/Steven_Rafael_Turner.pdf >> >> I am interested in machine translation because of its >> interdisciplinary nature. Linguistics is divided into several distinct >> areas: phonetics, phonology, morphology, syntax, semantics, >> pragmatics, etc. . Each of these areas has evolved throughout the >> years by developing its own ideas and techniques, and has reached by >> now a remarkable degree of specialization. Machine translation, >> however, requires that various connections between these areas and the >> creation of actual ‘bridges’ between different linguistic branches. >> Many times it seems that methods from all fields must employed to >> capture the nuance of language, and this interplay of different points >> of view in a same subject has played a fundamental role in >> illuminating the nature of language. Machine translation is a breeding >> ground for translating and comparing different linguistic theories, >> discovering analogies, and providing an effective means for >> transferring results and techniques between distinct fields. That is >> why I find machine translation so exciting! >> >> Please consider my prior experience in natural language processing, >> programming, and conceptualizing mathematically as proper >> qualifications. >> >> Take care, >> Rafael >> >> P.S. Please let me know if this is way out there. I have other ideas >> related to implementing morphology learning algorithms that may fit >> more naturally into the Apertium framework. > > Isn't this basically a character-based SMT system ? Remember that > Apertium is by-and-large an RBMT project -- have you looked over our > ideas on the ideas page?[1] > > Fran > > 1. http://wiki.apertium.org/wiki/Ideas_for_Google_Summer_of_Code > > > ------------------------------------------------------------------------------ > Enable your software for Intel(R) Active Management Technology to meet the > growing manageability and security demands of your customers. Businesses > are taking advantage of Intel(R) vPro (TM) technology - will your software > be a part of the solution? Download the Intel(R) Manageability Checker > today! http://p.sf.net/sfu/intel-dev2devmar > _______________________________________________ > Apertium-stuff mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/apertium-stuff > ------------------------------------------------------------------------------ Enable your software for Intel(R) Active Management Technology to meet the growing manageability and security demands of your customers. Businesses are taking advantage of Intel(R) vPro (TM) technology - will your software be a part of the solution? Download the Intel(R) Manageability Checker today! http://p.sf.net/sfu/intel-dev2devmar _______________________________________________ Apertium-stuff mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/apertium-stuff
