Análise de Medidas de Similaridade Semântica na Tarefa de Reconhecimento de Implicação Textual (Analysis of Semantic Similarity Measures in the Recognition of Textual Entailment Task)[In Portuguese]

نویسندگان

  • David Feitosa
  • Vladia Pinheiro
چکیده

In this work, we present a feature-based approach to the RTE (Recognizing Text Entailment) task that verifies the similarity between two sentences including syntactic and semantic aspects. The selected features come from the winning work of the RTE task of the workshop ASSIN (Semantic Similarity Evaluation and Textual Inference) with some changes and addition of other semantic feature. The evaluation methodology consisted in replicating the task with the database used in the workshop, analyzing the results with and without the semantic features. Besides the numerical approach, we mention a symbolic one with its characteristics and limitations. Resumo. Neste trabalho, apresentamos uma abordagem baseada no uso de features para a tarefa de RTE (Recognizing Text Entailment) que verifica a similaridade entre duas frases incluindo aspectos sintáticos e semântico. As features selecionadas são oriundas do trabalho vencedor da tarefa de RTE do workshop ASSIN (Avaliação de Similaridade Semântica e Inferência Textual) com algumas alterações e adições de outra feature semântica por nós. A metodologia de avaliação consistiu em replicar a tarefa com a base de dados usada no workshop, analisando os resultados com e sem as features semânticas. Além da abordagem numérica, citamos uma simbólica com suas caracterı́sticas e limitações.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Mecanismo de Encadeamento de Notícias por Reconhecimento de Implicação Textual

Atualmente, o acesso à informação relevante é feito com uma simples consulta a um sistema de busca na Web. Em se tratando de notícias, além de um sistema de consulta para o acesso à informação, sites especializados sugerem ao usuário notícias relacionadas para leitura complementar. No entanto, estas notícias relacionadas nem sempre complementam o conteúdo da primeira notícia lida pelo usuário. ...

متن کامل

Methodology and Results for the Competition on Semantic Similarity Evaluation and Entailment Recognition for PROPOR 2016

In this paper, we present the methodology and the results obtained by our teams, dubbed Blue Man Group, in the ASSIN (from the Portuguese Avaliação de Similaridade Semântica e Inferência Textual) competition, held at PROPOR 2016. Our team’s strategy consisted of evaluating methods based on semantic word vectors, following two distinct directions: 1) to make use of low-dimensional, compact, feat...

متن کامل

Una aproximación al RTE y a la Tarea de Búsqueda de Implicación Textual usando Máquinas de Soporte Vectorial

This paper shows a Recognizing Textual Entailment System, and a sub-system that address the Textual Entailment Search Task. This system employs a Support Vector Machine classifier with a set of 32 features, which includes lexical and semantic similarity for both two-way and three-way classification tasks. Additionally, we show an approach to dealing with the problem of searching entailment in a...

متن کامل

Geração de features para resolução de correferência: Pessoa, Local e Organização (Feature Generation for Coreference Resolution: Person, Location and Organization) [in Portuguese]

This work aims at resolving coreference in Portuguese, focusing on categories of named entities Person, Location and Organization. The proposed method uses supervised learning. To this end, the use of features that assist in the correct classification of named entities is critical. The construction and refinement of these features are of great relevance to his task. The performance of many othe...

متن کامل

Análise Automática de Coerência Textual em Resumos Científicos: Avaliando Quebras de Linearidade (Automatic Analysis of Textual Coherence in Scientific Abstracts: Evaluating Linearity Breaks)

This paper presents an extension of the coherence analysis module that is part of the writing tool called SciPo, allowing it to automate the analysis of the coherence dimension called Linearity Break. The proposed implementation is based on a combination of the entity grid model and information from the rhetorical structure of scientific abstracts, allowing it to generate messages that indicate...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2017