By Hrafn Loftsson, Eirikur Rögnvaldsson, Sigrun Helgadottir
This publication constitutes the complaints of the seventh overseas convention on Advances in common Language Processing held in Reykjavik, Iceland, in August 2010.
Read Online or Download Advances in Natural Language Processing: 7th International Conference on NLP, IceTAL 2010, Reykjavik, Iceland, August 16-18, 2010, Proceedings PDF
Best compilers books
Initially released in 1981, this was once the 1st textbook on programming within the Prolog language and remains to be the definitive introductory textual content on Prolog. notwithstanding many Prolog textbooks were released due to the fact that, this one has withstood the attempt of time due to its comprehensiveness, educational process, and emphasis on common programming purposes.
- Algorithm Design For Networked Information Technology Systems
- Einführung in die Programmierung : Grundlagen, Java, UML
- The Structure of the Relational Database Model
- Automatic Quantum Computer Programming: A Genetic Programming Approach
- An introduction to language processing with Perl and Prolog : an outline of theories, implementation, and application with special consideration of English, French, and German
Extra info for Advances in Natural Language Processing: 7th International Conference on NLP, IceTAL 2010, Reykjavik, Iceland, August 16-18, 2010, Proceedings
Distractors: The distractors were automatically generated for each blank and method. In the case of the nouns, the four mentioned methods were applied and in the case of the verbs, two methods were applied: the LSA method and the LSA & specialised dictionary method8 . As the distractor generation task is completely automatic, the possibility of generating distractors that are correct in the given context had to be considered. That is why before testing them with learners the distractors were manually checked.
C Springer-Verlag Berlin Heidelberg 2010 28 I. Aldabe and M. Maritxalar methods to automatically generate distractors of high quality. That is to say, distractors that correspond to the vocabulary studied by learners as part of the curricula. As there must be only one possible answer among the options of each MCQ, experts had to discard those distractors that could form a correct answer. Our purpose was to evaluate the system itself by means of an evaluation in a real situation with learners. The results of a test exercise was used to measure the quality of the automatically generated distractors.
Word related features: the highly informative features described above. They consist of a large set of binary features indicating if the word belongs or not to a speciﬁc subdictionary. For example, if the word beautiful appears in the nouns, adjectives and compliments dictionaries, the corresponding binary features are set to true. This implies that the word can be classiﬁed as noun, adjective or as the semantic tag compliments. – Sentence related features: introduced to support the diﬀusion of relevant pieces of information along the whole sentence.