The Computational Linguistics - Applications Workshop was created in 2008 in response to the fast-paced progress in the area.
Traditionally, computational linguistics was limited to the scientists specialized in the processing of a natural language by computers. Scientific approaches and practical techniques come from linguistics, computer science, psychology, and mathematics. Nowadays, there is a number of practical applications available. These applications are sometimes developed by smart yet NLP-untrained developers who solve the problems using sophisticated heuristics.
Computational Linguistics needs to be applied to make the full use of the Internet. There is a definite need for software that can handle unstructured text to allow search for information on the web. According to the European Commission, Human Language Technologies are one of the key research areas for the upcoming years. The priority aim of the research in this area is to enable users to communicate with the computer in their native language.
CLA'10 Workshop is a place where the parties meet to exchange views and ideas with a benefit to all involved. The Workshop will focus on practical outcome of modeling human language use and the applications needed to improve human-machine interaction.
This call is for papers that present research and developments on all aspects of Natural Language Processing used in real-life applications, such as (this list is not exhaustive):
- information retrieval
- extraction of linguistic knowledge from text corpora
- semantic ontologies in computer linguistics
- lexical resources
- machine translation and translation aids
- ambiguity resolution
- text classification
- corpus-based language modeling
- parsing issues
- proofing tools
- dialogue systems
- machine learning methods applied to language processing
- ontology and taxonomy evaluation
- opinion mining
- question answering
- sentiment analysis
- speech and audio processing
- text summarization
- use of NLP techniques in practical applications
The presentation of the paper has to include a demonstration of an existing tool. The papers should include a section describing the tool (or a prototype), which demonstrates the theory discussed in the paper.
The presentation is divided into two parts. First, the author(s) shortly demonstrate their tools to the audience. In the second part, the authors discuss their work with other participants and let the audience test their software.
Papers will be evaluated and accepted on the basis of their technical merit, usefulness of the real life application and relevance to the workshop scope by the CLA'10 Program Committee. The paper will be assessed by academics as well as industry representatives in order to assure fair and balanced assessment.
All accepted and presented papers will be published in the Conference Proceedings and included in the IEEE Xplore® database. The best demonstrations will be selected to be shown to the general audience of the conference at a plenary session.
- Authors should submit draft papers (as Postscript, PDF of MSWord file).
- The total length of a paper should not exceed 8 pages (IEEE style). IEEE style templates will be available ASAP.
- Papers will be refereed and accepted on the basis of their scientific merit and relevance to the workshop.
- Accepted and Presented paper will be published in the Conference Proceedings and included in the IEEE Xplore® database.
- Organizers reserve right to move accepted papers between IMCSIT events.
31.05.2010 (May 31, 2010) - Full paper submission
12.07.2010 (July 12, 2010) - Notification of acceptance
23.08.2010 (August 23, 2010) - Camera-ready version of the accepted paper
CLA '09 post conference report
It seems that the motto of our workshop "Show the system in action" has been accepted by the NLP community. This year 33 papers were submitted (90% increase from last year); 19 of them were accepted by the programming committee. 15 papers were presented during the workshop - our guests from Iran or Tunisia seemingly found Poland too far away from their homelands. Still, 11 countries were represented in the workshop, mainly from the Western Europe.
The workshop was composed of oral presentations and the demo-poster session. Presentations, given on Tuesday (October, 12th), were divided into three sessions:
- Information Retrieval,
- Use of NLP Tools In practical application,
- Parsing issues; Linguistic resources.
In the evening the audience voted for the best presentations of each of the session as well as the best overall performance. The voters used their mobile phones and the result were known immediately - to the chairman. The winners were presented officially at the Gala Ceremony (In the Oscar Ceremony fashion). The Information Retrieval session was one by Igor Leturia from Spain - or rather his virtual agent who could fluently answer any question posed in the Basque language. The session on Practical Applications was won by Eric Wehrli from Switzerland and his micro-scanner that read German texts and translated them into English. Parsing issues were best presented by charming Katarzyna Moroz from Poznan (Poland) who convinced the audience that her new parsing algorithm using pregroup grammars will soon outperform existing ones. And finally, after the three nominees had been presented, the Oscar went to...Spain.
The demo-poster session took place on Wednesday morning. It looked as if the participants had not celebrated too long into the Tuesday night - at 9 a.m. almost all of them were ready to put their applications under the pressure of real tests. Well, the winner did not understand the chairman's Basque too well... she obviously preferred her master, Igor.