UNED submission to AVE 2006

2007

This paper describes the INAOE's answer validation system evaluated at the Spanish track of the AVE 2007. This system is based on a supervised learning approach that considers two kinds of attributes. On the one hand, some attributes indicating the textual entailment between the given support text and the hypothesis constructed from the question and answer. On the other hand, some new features denoting certain answer restrictions as imposed by the question's type and format. In order to extract all these attributes the system uses different tools such as a lemmatizer, a POS tagger, a NER procedure and a superficial syntactic parser. Experimental results are encouraging; they show that the proposed system achieved a 52.91% of F-measure and that it outperformed the standard baseline by 15 percentage points.

2006

Information and Communication Technologies Structure of the talk A light introduction to NL(P) Very brief presentation of Linguateca Evaluation contests Named entity recognition HAREM 3 Information and Communication Technologies What is natural language (processing)? Natural language is the oldest and most successful knowledge representation language Used for comunication, negotiation, and reason (->logic) Main features:

Accessing Multilingual …, 2006

This paper describes the TALP-QA system in the context of the CLEF 2005 Spanish Monolingual Question Answering (QA) evaluation task. TALP-QA is a multilingual open-domain QA system that processes both factoid (normal and temporally restricted) and definition questions. The approach to factoid questions is based on in-depth NLP tools and resources to create semantic information representation. Answers to definition questions are selected from the phrases that match a pattern from a manually constructed set of definitional patterns.

2012

strong links among the linguistic resources developed for each language of the region is essential. Additionally, a lack of published resources in some of these languages exists. Such lack propitiates a strong inter-relation between them and higher resourced languages, such as English and Spanish. In order to favour the intra-relation among the peninsular languages as well as the inter-relation between them and foreign languages, different purpose multilingual NLP tools need to be developed. Interesting topics to be researched include, among others, analysis of parallel and comparable corpora, development of multilingual resources, and language analysis in bilingual environments and within dialectal variations. With the aim of solving these tasks, statistical, linguistic and hybrid approaches are proposed. Therefore, the workshop addresses researchers from different fields of natural language processing/computational linguistics: text mining, machine learning, pattern recognition, i...

2011

Text adaptation is a normal activity of teachers to facilitate reading comprehension of specific contents; the general approaches for it are Text Simplification and Text Elaboration (TE). TE aims at clarifying, explaining information and making connections explicit in texts. In this paper, we present a new approach for TE: an automatic question categorization system which assigns wh-question labels to verbal arguments in a sentence. For example, in “Mary danced yesterday.” “Who?” is the label linking the verb “danced” to the argument “Mary” and “When?” links “danced” to the argument “yesterday”. This annotation is similar to semantic role labeling, approached successfully via statistical language processing techniques. Specifically, we present experiments to build the system using a fine-grained question set in Portuguese language and address two key research questions: (1) Which machine-learning algorithm presents the best results? (2) Which problems this task presents and how to o...

Lecture Notes in Computer Science, 2005

We present in this work a method for question classification in Spanish and Portuguese. The method relies on lexical features and attributes extracted from the Web. A machine learning algorithm, namely Support Vector Machines is successfully trained on these features. Our experimental results show that this method performs consistently well over two different languages.

International Journal of Future Computer and Communication, 2013

2011

Preface In the Iberian Peninsula, five official languages co-exist: Basque, Catalan, Galician, Portuguese and Spanish. Fostering multi-linguality and establishing strong links among the linguistic resources developed for each language of the region is essential. Additionally, a lack of published resources in some of these languages exists. Such lack propitiates a strong interrelation between them and higher resourced languages, such as English and Spanish. In order to favour the intra-relation among the peninsular languages as well as the interrelation between them and foreign languages, different purpose multilingual NLP tools need to be developed. Interesting topics to be researched include, among others, analysis of parallel and comparable corpora, development of multilingual resources, and language analysis in bilingual environments and within dialectal variations. With the aim of solving these tasks, statistical, linguistic and hybrid approaches are proposed. Therefore, the wor...

Anais do XXIX Simpósio Brasileiro de Informática na Educação (SBIE 2018)

Automatic Short Answer Grading is the study field that addresses the assessment of students' answers to questions in natural language. Besides length, it differs from automatic essay grading by focusing on the evaluation of content instead of answer's style. The grading of the answers is generally seen as a typical classification supervised learning. Many works have been recently developed, but most of them deal with data in the English language. In this paper, we present a new Portuguese dataset and system for automatic short answer grading. The data was collected with the participation of 13 teachers, 12 undergraduate students and 245 elementary school students. Results achieved 69% accuracy in four-class classification and 85% on binary classification.

Lecture Notes in Computer Science, 2010

In previous work we have proved that the BLEU algorithm , originally devised for evaluating Machine Translation systems, can be applied to assessing short essays written by students. In this paper we present a comparative evaluation between this BLEU-inspired algorithm and a system based on Latent Semantic Analysis. In addition we propose an effective combination schema for them. Despite the simplicity of these shallow NLP methods, they achieve state-of-theart correlations to the teachers' scores while keeping the language-independence and without requiring any domain specific knowledge.