Irene Pimenta Rodrigues

A logic programming based approach to QA@CLEF05 track

In this paper the methodology followed to build a question-answering system for the Portuguese language is described. The system modules are built using computational linguistic tools such as: a Portuguese parser based on constraint grammars for the syntactic analysis of the documents sentences and the user questions; a semantic interpreter that rewrites sentences syntactic analysis into discourse representation structures in order to obtain the corpus documents and user questions semantic representation; and finally, a semantic/pragmatic interpreter in order to obtain a knowledge base with facts extracted from the documents using ontologies (general and domain specific) and logic inference. This article includes the system evaluation under the CLEF’05 question and answering track.

Using logic programming to model Multi-Agent web legal systems – an application report

A logic programming framework for the definition of cooperative multi-agent legal web information retrieval systems is proposed. Cooperation is achieved through the use of dialogue processing techniques, namely, the inference of the user intentions and the existence of a pro-active system behaviour, which tries to help users in their searches. The proposed architecture has a core IR module, which accesses the legal knowledge bases, and three specialised logic programming agents: an agent manager that receives the user web initial requests and it establishes the connection with a specific user agent; a user agent, which is specific to each user, and it has information about the user profile and the previous interrogation context; and an agent monitor that informs the agent manager of the latest changes in the knowledge bases allowing these changes to be transmitted to all users which have one of their previous queries results changed. The logic programming modules were defined using dynamic logic programming and LUPS, a language for updates [1, 3]. The proposed framework was implemented in a Linux environment using XSB Prolog and it was applied to the legal knowledge base of the Portuguese Attorney General [15]. The evaluation results show that the integration of dialogue processing techniques with a legal IR system, allow an improvement of the system, namely, decreasing the average number of interactions per query.

PGR: portuguese attorney general's office decisions on the web

A multi-agent based architecture for the Portuguese Attorney Generals Office documents is presented. The architecture has two layers: the first one uses natural language processing techniques to manage the legal text bases; the second one, uses dynamic logic programming to define agents, which cooperatively handle the interaction between users and the text bases. The natural language processing layer uses a lexical dictionary and a Portuguese POS (part of speech) tagger to improve the results of the text search engine. Moreover, the retrieved documents are clustered in topics, helping the users to refine their queries and to select the desired documents. The interaction layer uses agents to model cooperativity and to handle the multi-modal interactions (natural language sentences and graphical actions). The proposed architecture was implemented in a Linux environment using Prolog.

A collaborative legal information retrieval system using dynamic logic programming

We propose a framework for a collaborative legal information retrieval system based on dynamic logic programming. In order to be collaborative our system keeps the context of the interaction and tries to infer the user intentions. Each event is represented by logic programming facts which are used to dynamically update the previous user model. User intentions are inferred from this new model and are the basis of the interaction between the system and the legal texts knowledge base. As legal texts knowledge base we are using the documents from the Portuguese Attorney General (Procuradoria Geral da República). In this paper we will show some examples of the obtained collaborative behaviour.

Work out the semantic web search: the cooperative way

We propose a Cooperative Question Answering System that takes as input natural language queries and is able to return a cooperative answer based on semantic web resources, more specifically DBpedia represented in OWL/RDF as knowledge base and WordNet to build similar questions. Our system resorts to ontologies not only for reasoning but also to find answers and is independent of prior knowledge of the semantic resources by the user. The natural language question is translated into its semantic representation and then answered by consulting the semantics sources of information. The system is able to clarify the problems of ambiguity and helps finding the path to the correct answer. If there are multiple answers to the question posed (or to the similar questions for which DBpedia contains answers), they will be grouped according to their semantic meaning, providing a more cooperative and clarified answer to the user.

A Semantic Web Pragmatic Approach to Develop Clinical Ontologies, and Thus Semantic Interoperability, Based in HL7 v2.XML Messaging

The ISO/HL7 27931:2009 standard intends to establish a global interoperability framework for Healthcare applications. However, being a messaging related protocol, it lacks a semantic foundation for interoperability at a machine treatable level has intended through the Semantic Web. There is no alignment between the HL7 V2.xml message payloads and a meaning service like a suitable ontology. Careful application of Semantic Web tools and concepts can ease extremely the path to the fundamental concept of Shared Semantics. In this paper the Semantic Web and Artificial Intelligence tools and techniques that allow aligned ontology population are presented and their applicability discussed. We present the coverage of HL7 RIM inadequacy for ontology mapping and how to circumvent it, NLP techniques for semi automated ontology population and discuss the current trends about knowledge representation and reasoning that concur to the proposed achievement.

A question-answering system for Portuguese juridical documents

We present a Question-Answering (QA) system for Portuguese juridical documents.The QA system was applied to the complete set of decisions from several Portuguese juridical institutions (Supreme Courts, High Court, Courts, and Attorney-Generals Office) in a total of 180,000 documents.

A natural language interface for information retrieval on semantic web documents

We present a dialogue system that enables the access in natural language to a web information retrieval system. We use a Web Semantic Language to model the knowledge conveyed by the texts. In this way we are able to obtain the associated knowledge necessary to perform the different analysis stages of natural language sentences. In the context of information retrieval, we aim to develop a system that, by increasing the interaction management capabilities, is able to achieve a better degree of cooperativeness and to reduce the average number of interactions needed to retrieve the intended set of documents. The documents in the IR system considered here are composed by the set of documents produced by the Portuguese Attorney General since 1940. These documents were analyzed and an ontology describing their structure and content was defined. Then, they were automatically parsed and a (partial) semantic structure was created. The ontology and the semantic content was represented in the OWL language. An example of a user interaction session is presented and explained in detail.

A multi-agent approach to question answering

In this paper we present a multi-agent approach to question answering for the Portuguese language. Our proposal is composed by three modules: (1) document and query processing; (2) ontology construction; and (3) answer generation. Each module is composed by multiple cooperative agents which adopt distinct strategies to generate its outputs and cooperate to create a global result. This approach allows the use of different strategies and the reduction of errors introduced by individual methods. The cooperation among the agents aims to reach better solutions in each step of the processing.

Using dialogues to access semantic knowledge in a web IR system

We present a dialogue system that enables the access in natural language to a web law information retrieval system. We use a semantic web language to model the document knowledge and to define an ontology representing the main classes of domain objects, their properties and their relations. Our system includes an ontology in DAML+OIL language describing the documents structure, a document database build with a subset of the Portuguese Attorney Generals Office documents were each document includes a field with its semantic content described in DAML+OIL. The dialogue system interprets the user natural language sentences using the ontology, the documents semantic content and the inferred user attitudes.