Winners in 2010 NACLO competition
« previous post | next post »
The winners in the Fourth Annual North American Computational Linguistics Competition have been announced. The top eight scorers were:
1st- Ben Sklaroff, Palo Alto, CA, Palo Alto High School
2nd- Brian Kong, Milton, MA, Milton Academy
3rd- Allen Yuan, Farmington Hills, MI, Detroit Country Day School
4th- Daniel Li, Fairfax, VA, Thomas Jefferson High School for Science and Technology
5th- Alan Chang, San Jose, CA
6th- Alexander Iriza, Astoria, NY, The Dalton School
7th- In-Sung Na, Old Tappan, NJ, Northern Valley Regional HIgh School at Old Tappan
8th- Tian-Yi Damien Jiang, Raleigh, NC, North Carolina School of Science & Mathematics (Durham)
1,118 students participated in the first round of the competition, held on February 4, and the top 100 scorers took part in a second round on March 10. Squads formed from the best-scoring participants will be eligible to go on to the Eighth International Linguistics Olympiad, to be held in Sweden in July. NACLO winners have done very well in previous Olympiads.
According to the press release,
Students compete in the Computational Linguistics Olympiad by solving challenging problems using data from a variety of languages and formal systems. There is no pre-requisite knowledge. Students discover facts about languages and formal systems in the course of solving the puzzles. This year students solved a total of 16 problems, including deciphering the rules for a Pig-Latin-like play language in Minangkabau, the writing systems of Plains Cree, and the Vietnamese classic Tale of Kieu written in Chinese characters. Computational problems dealt with text compression and automatic expansion of abbreviated words.
For those interested in more detail, the NACLO website gives the problems from previous years. For example, the first problem in the invitational round of the 2009 competition involved figuring out a shift-reduce parser for the button presses in P-Little's Triple-I XTreem Hyp()th3tica7 Sk8boarding Game; the second one involves decoding Linear B; and the third one requires you to figure out some aspects of Bulgarian morphology.
The NACLO website also explains the history:
The idea of holding academic Challenges in linguistics stems from a long tradition of linguistics and mathematics competitions, which began in Moscow in the 1960s. In 1984, Bulgaria began holding similar competitions, and contests were first held in the United States at the University of Oregon starting in 1998. Bulgaria hosted the First International Olympiad in Linguistics in Borovetz in September of 2003, and subsequent International Olympiads have been held in Moscow, Russia in 2004, Leiden, The Netherlands in 2005, and in Tartu, Estonia in 2006. More recently, universities in Estonia, Finland, Netherlands, the United States, and other countries have begun sponsoring such outreach activities aimed at high school students. Participating as individuals and in country teams, students are given challenging sets of language data and language puzzles to solve, with the chance to win prizes and international recognition. Students learn about the richness, diversity and systematicity of language, while exercising natural logic and reasoning skills. No prior knowledge of languages or linguistics is necessary, but the competitions have proven very successful in attracting top students to study in the field of linguistics and computational linguistics.
The North American Computational Linguistics Olympiad picks up on this long tradition, with a focus on computational thinking as it relates to solving linguistics problems. In addition to the traditional linguistics problems, NAMCLO endeavors to introduce students to computational problem solving as it relates specifically to natural language data.
The people mainly responsible for creating this excellent institution are Lori Levin and Drago Radev.
Nat said,
April 3, 2010 @ 7:12 pm
Why not give them Linear A? A missed opportunity.
Bob said,
April 3, 2010 @ 7:25 pm
I love these problem sets! A much nicer way to pass the weekend than a crossword puzzle.
[(myl) Maybe some enterprising on-line magazine should start publishing linguistic puzzles as a feature.]
Will said,
April 3, 2010 @ 9:56 pm
Looking through some of the problem sets from previous years–and having fun solving them–it's evident that while no prerequisite linguistic knowledge is necessary (as claimed), a little bit of such knowledge definitely goes a long way. Just knowing a few simple things like syllable structure and phonetic composition made solving certain problems much easier. I wonder if the students entering this competition tend to have some basic training of linguistics, or if they just go in with critical thinking and hope for the best. Then again, I guess part of the purpose of the competition is to introduce students to linguistic concepts by a sort of problem-set analogue to the Socratic method.
David Cantor said,
April 4, 2010 @ 4:05 am
Interesting to note that 6 out of the 8 top scorers have what appear to be Asian surnames. Other top math and science competitions have very similar patterns. This is why I cringe when I hear politicians who want to limit or ban immigration.
john riemann soong said,
April 4, 2010 @ 6:55 pm
Seems like 7/8 to me …
Which is weird cuz it seems like I'm the only Asian at my school interested in linguistics, never mind psycholinguistics/language acquisition.
J.W. Brewer said,
April 4, 2010 @ 11:15 pm
The current immigrant (in the sense of foreign-born) population of the United States includes approx. twice as many persons born in Latin America as in Asia (see, e.g., Table 44 in the Statistical Abstract of the U.S.). Yet Mr. Cantor may have noticed that Spanish-surnamed winners of math and science competitions are not typically twice as common as Asian-surnamed winners (or some more precise ratio if the age distributions are different, which I haven't bothered to look up). Which of course isn't to say that our immigration policy should be determined either in part or in whole by math-and-science-competition aptitude, only that it shouldn't be determined by anecdote and romanticization. (Table 53, on the numbers of people speaking specified non-English languages at home, may also be interesting to consider in this context — but note the categorization methodology there in which neither languages indigenous to the Indian subcontinent nor Tagalog are deemed "Asian" languages although their speakers tend to be lumped into the "Asian-American" category in most U.S. bureaucratic/demographic contexts.)
uberVU - social comments said,
April 5, 2010 @ 2:57 pm
Social comments and analytics for this post…
This post was mentioned on Twitter by PhilosophyFeeds: Language Log: Winners in 2010 NACLO competition http://goo.gl/fb/Blcjm…
Supergrunch said,
April 6, 2010 @ 7:48 pm
I remember doing one of the 3 hour papers for fun the summer before I started my degree in linguistics. Was a great introduction to the subject, I think puzzles are really the way to introduce people to the field (as in Richard Hudson's Invitation to Linguistics, which I read around the same time). Whenever I get tangled up in theoretical conflicts it's comforting to think of how straightforward and analytic things can be.
Alex Iriza said,
April 13, 2010 @ 7:10 pm
Just for the record, I'm not asian :) No idea where Iriza comes from though…
Peter Bakker said,
April 15, 2010 @ 3:54 am
Alex, your name may be Basque if it was originally Irizar. Lots of buses in Europe are produced by this company. In that case it would probably mean "old city": iri = city, zahar (dialectal zar) = old.