Vassilis Konstantinou

Can legal knowledge be derived from legal texts

Knowledge acquisition is undoubtedly one of the major bottle-necks in the development of legal expert systems. Usually the knowledge is collected by knowledge engineers who are forced to make their own interpretations of the knowledge in order to map it on a knowledge representation technique, thus resulting into erroneous and legally unacceptable interpretations of the law. The aim of NOMOS (an EC supported project under the ESPRIT II initiative) was to assist the knowledge engineer by providng tools that perform semi-automatic knowledge acquisition from legal texts in Italian and French. This paper reports on the results of the first evaluation of the knowledge collected by these tools. The evaluation was performed by complementing the tools with a fully functional expert system that accepted the generated knowledge bases and allowed experts to test the completeness of the knowledge through a series of interactive consultations. The knowledge base used for this evaluation was derived from the text for the Italian Value Added Tax Law. The text was pre-processed in its ASCII form by the Nomos tools and the generated knowledge base was filtered through to a conventional expert system shell to generate the evaluation expert system. Knowledge extracted directly from text was converted into a hybrid of production rules and Conceptual Graphs. [see SOWA 1984] Knowledge collected from other sources, such as previously resolved cases, explanations of terms and examples, were linked to the knowledge base using an automated hypertext technique. [see KONSTANTINOU & MORSE 1992] Finally, the expert system was tested using real-life cases supplied by the Italian ministry of finance.

A Dynamic JAVA-Based Intelligent Interface for Online Image Database Searches

This paper describes the use of a JAVA-based interface to enable remote image-based searches. The database used for our target system, resides at the Marcianna Library in Venice, Italy. It consists of a number of medieval manuscripts scanned at a variety of resolutions and indexed by both textual and image based information. The system described in this paper is a JAVA based client-server which can be used to interrogate a remote hybrid database and initiate searches using both the textual and image links of this database. Image queries can be based on data supplied by the user, or extracts of the original remote database. In the paper we also discuss the indexing and pattern recognition algorithms which are used on the Server side. This work is part of the HISTORIA project funded under the European Union Libraries Initiative (Telematics for Libraries - Project No:3117).

Ink Discrimination Based on Co-occurrence Analysis of Visible and Infrared Images

Inks found in Byzantine manuscripts are semi- transparent pigments and their examination and analysis provide an invaluable source of information on the authenticity and dating of manuscripts and the number of authors involved. However, inks are difficult to characterize because their intensity depends on the amount of liquid spread during scripting and the reflective properties of the support. Most existing methods for the analysis of ink materials are based on destructive testing techniques that require the physicochemical sampling of data. Such methods cannot be widely used because of the historical and cultural value of the manuscripts. In this work we show that manuscript inks can be represented through a mixture of Gaussian functions and can be characterised using co-occurrence matrices.

Legal reasoning methodology: The missing link

Abstract The biggest false assumption made when attempting to model automatically acquired legal knowledge is that methodological and procedural legal knowledge is also contained in the text of law. Although the legal profession Intuitively knows the falsity of this assumption, researchers in the area of automatic knowledge acquisition are still confident in implementing systems that use only the text of laws as their main source of knowledge. Knowledge engineers are then forced to make their own interpretations of this knowledge, thus resulting in erroneous and legally unacceptable interpretations of the law. The aim of Nomos (an EC supported project under the ESPRIT II initiative) was to assist the knowledge engineer by providing tools that perform semiautomatic knowledge acquisition from legal texts in Italian and French. This paper uses the implementation of Nomos‐advisor, a legal expert system that uses Nomoss results as an input, as a proof of the falsity of the above assumption and discusses possi...

Electronic documentation system: using automated hypertext techniques for technical support services

The Electronic Documentation System, was developed to provide fast, easy, online information retrieval, meeting a specific need for engineering industries. It addresses the requirements of both the user and the author of electronic documentation and provides a viewing system that can be used across a variety of platforms. It also takes advantage of sophisticated cross-referencing techniques to automate the development of electronic documents.

Content-based search in medieval manuscripts

HISTORIA is one of the six projects funded by the European Union Telematics in Libraries programme. The objective of the project is the design and realisation of a digital image library equipped with search and query tools for coding and retrieving information from reproductions of medieval manuscripts that contain heraldic images. In this paper we discuss the particular problems encountered in the indexing and search of manuscripts related to the geneology of Venetian noble families and their associated heraldry and present a prototype digital library.

A Bayesian decision model for watercolour analysis

Bayesian Classification methods can be applied to images of watercolour paintings in order to characterize blue and green pigments used in these paintings. Pigments found in watercolour paintings are semi-transparent materials and their analysis provides important information on the date, the painter, the place of the production of watercolour paintings and generally on the authenticity of these works of art. However, watercolour pigments are difficult to characterize because their intensity depends on the amount of liquid spread during painting and the reflective properties of the underlying support. The method describedin this paper is non-destructive, non invasive, does not involve sampling and can be applied in situ. The methodology is based on the photometric properties of pigments and produce computational models which classify diverse types of pigments found in watercolour paintings. These pigments are photographed in the visible and infrared area of electromagnetic spectrum and models based on statistical characteristics of intensity values using a mixture of Gaussian functions are created. Finally the pigments are classified using a Bayesian classification algorithm to process the generate models.

Probabilistic image-based characterization of manuscript inks

Inks constitute the main element in Medieval manuscripts and their examination and analysis provides an invaluable source of information on the authenticity of the manuscripts, the number of authors involved and dating of the manuscripts. Most existing methods for the analysis of ink materials are based on destructive testing techniques that require the physicochemical sampling of data. Such methods cannot be widely used because of the historical and cultural value of manuscripts. In this paper we present a novel approach for discriminating and identifying inks based on the correlations of image variations under visible and infrared illumination. Such variations are studied using co-occurrence matrices and detect the behavior of the inks during the scripting process.

Extracting Explicit and Implicit Knowledge from Natural Language Texts

The use of conceptual graphs in information retrieval and natural language applications has grown rapidly in popularity and sophistication over the past few years. Question answering/retrieval systems

Postcards: hypermedia for the travel industry

In this paper we will discuss the development of a hypermedia application called Postcards, for the travel industry. Postcards is an interactive travel guide to California that incorporates digital video, animation and graphics with a highly flexible hypertext engine. The main objectives of this project were to createaninterface and set of tools that would increase usability, and reduce disorientation; to test if the inclusion of the muldmedia elements would actually improve the usability and functionality of the information; and finally to produce a commercially viable product.