José Saias

A Methodology to Create Legal Ontologies in a Logic Programming Based Web Information Retrieval System

Web legal information retrieval systems need the capability to reason with the knowledge modeled by legal ontologies. Using this knowledge it is possible to represent and to make inferences about the semantic content of legal documents. In this paper a methodology for applying NLP techniques to automatically create a legal ontology is proposed. The ontology is defined in the OWL semantic web language and it is used in a logic programming framework, EVOLP+ISCO, to allow users to query the semantic content of the documents. ISCO allows an easy and efficient integration of declarative, object-oriented and constraint-based programming techniques with the capability to create connections with external databases. EVOLP is a dynamic logic programming framework allowing the definition of rules for actions and events. An application of the proposed methodology to the legal web information retrieval system of the Portuguese Attorney General’s Office is described.

A methodology to create legal ontologies in a logic programming information retrieval system

Legal web information retrieval systems need the capability to reason with the knowledge modeled by legal ontologies. Using this knowledge it is possible to represent and to make inferences about the semantic content of legal documents. In this paper a methodology for applying NLP techniques to automatically create a legal ontology is proposed. The ontology is defined in the OWL semantic web language and it is used in a logic programming framework, EVOLP+ISCO, to allow users to query the semantic content of the documents. ISCO allows an easy and efficient integration of declarative, object-oriented and constraint-based programming techniques with the capability to create connections with external databases. EVOLP is a dynamic logic programming framework allowing the definition of rules for actions and events. An application of the proposed methodology to the legal information retrieval system of the Portuguese Attorney Generals Office is described.

Sentiue: Target and Aspect based Sentiment Analysis in SemEval-2015 Task 12

This paper describes our participation in SemEval-2015 Task 12, and the opinion mining system sentiue. The general idea is that systems must determine the polarity of the sentiment expressed about a certain aspect of a target entity. For slot 1, entity and attribute category detection, our system applies a supervised machine learning classifier, for each label, followed by a selection based on the probability of the entity/attribute pair, on that domain. The target expression detection, for slot 2, is achieved by using a catalog of known targets for each entity type, complemented with named entity recognition. In the opinion sentiment slot, we used a 3 class polarity classifier, having BoW, lemmas, bigrams after verbs, presence of polarized terms, and punctuation based features. Working in unconstrained mode, our results for slot 1 were assessed with precision between 57% and 63%, and recall varying between 42% and 47%. In sentiment polarity, sentiue’s result accuracy was approximately 79%, reaching the best score in 2 of the 3 domains.

A Methodology to Create Ontology-Based Information Retrieval Systems

Modern information retrieval systems need the capability to reason about the knowledge conveyed by text bases.

Senti.ue: Tweet Overall Sentiment Classification Approach for SemEval-2014 Task 9

This document describes the senti.ue system and how it was used for participation in SemEval-2014 Task 9 challenge. Our system is an evolution of our prior work, also used in last year’s edition of Sentiment Analysis in Twitter. This system maintains a supervised machine learning approach to classify the tweet overall sentiment, but with a change in the used features and the algorithm. We use a restricted set of 47 features in subtask B and 31 features in subtask A. In the constrained mode, and for the five data sources, senti.ue achieved a score between 78,72 and 84,05 in subtask A, and a score between 55,31 and 71,39 in subtask B. For the unconstrained mode, our score was slightly below, except for one case in subtask A.

The University of Évora's Participation in QA@CLEF-2007

The University of Evora participation in QA@CLEF-2007 was based on the Senso question answer system. This system uses an ontology with semantic information to support some operations. The full text collection is indexed and for each question a search is performed for documents that may have one answer. There is an ad-hoc module and a logic-programming based module that look for answers. The solution with the highest weight is then returned. The results indicate that the system is more suitable for the definition question type.

University of Évora in QA@CLEF-2004

The approach followed by the University of Evora team when building a system for participation in the CLEF 2004 question answering task for Portuguese is described. The system is based on two steps: for each question, a first search selects a set of potentially relevant documents; each of these documents is then analysed to obtain a semantic representation and the answer to the initial query. This approach was applied to the QA@CLEF test set for Portuguese with interesting results that have allowed us to identify the strong and weak features of our system.

Information extraction for standardization of tourism products

Tourism product descriptions are strongly supported on natural language expressions. Appropriate offer selection, according to tourist needs, depends highly on how these are communicated. Since no human interaction is available while presenting tourism products online, the way these are presented, even when using only textual information, is a key success factor for tourism web sites to achieve a purchase. Due to the large amount of tourism offers and the high dynamics in this sector, manual data management is not a reliable or a scalable solution. This paper presents a prototype developed for automatic extraction of relevant knowledge from tourism-related natural language texts. Captured knowledge is represented in a normalized format and new textual descriptions are produced according to available marketing channels. At this phase, the prototype is focused on hotel descriptions and is already using real operational data retrieved from the KEY for Travel tourism platform.

A Proposal for a Web Information Extraction and Question-Answer System

The Web is part of today’s life and offers all kind of content. We present a system that can help the user to extract information from web documents and to find the answer for simple questions in natural language. This work is focused on newspaper articles and it is based on an ontology knowledge representation, natural language processment and a logic-programming framework.