Eurovoc Thesaurus Research Articles

Automated subject indexing using word embeddings and controlled vocabularies: a comparative study

Text mining methods contribute significantly to the understanding and the management of digital content, increasing the potential of entry links. This paper introduces a method for subject analysis combining topic modelling and automated labelling of the generated topics exploiting terms from existing knowledge organisation systems. A testbed was developed in which the Latent Dirichlet Allocation (LDA) algorithm was deployed for modelling the topics of a corpus of papers related to the Digital Library Evaluation domain. The generated topics were represented in the form of bags-of-words word embeddings and were utilised for retrieving terms from the EuroVoc Thesaurus and the Computer Science Ontology (CSO). The results of this study show that the domain of DL can be described with different vocabularies, but during the process of automatic labelling the context needs to be taken into account.

Policyminer: From Oysters to Pearls

Today, a historically unprecedented volume of data is available in the public domain with the potential of becoming useful for researchers. More than at any other time before, political parties and governments are making data available such as speeches, legislative bills and acts. However, as the size of available data increases, the need for sophisticated tools for web-harvesting and data analysis simultaneously grows. Yet, for the most part researchers who are developing these tools come from a computer science background, while researchers in the social and behavior sciences who have an interest in using such tools often lack the necessary training to apply these tools themselves.In order to provide a bridge between these two communities we propose a new tool called PolicyMiner. The objective of this tool is twofold: First, to provide a general purpose web-harvesting and data clean-up tool which can be used with relative ease by researchers with limited technical backgrounds. The second objective is to implement knowledge discovery algorithms that can be applied to textual data, such as legislative acts. With our paper we present a technical document which details the steps of data processing that have been implemented in the PolicyMiner. First, the PolicyMiner harvests the raw html data from publically available websites, such as governmental sites, and provides a unique integrated view for the data. Second, it cleans the data by removing irrelevant items, such as html tags and non-informative terms. Third, it classi es the harvested data according to a pre-de ned standard conceptual hierarchy relying on the Eurovoc thesaurus. Fourth, it applies di fferent knowledge discovery algorithms such as time series and correlation-based analysis to capture the temporal and substantive policy dependencies of the textual data across countries.

El <i>Diari Oficial de la Generalitat de Catalunya</i>: Génesis de un Diario Digital

The entry into force of the Spanish act about digital access (Ley 11/2007, BOE n. 150, de 23/6/2007) has produced a decisive impact on the development of records management projects in organizations. The Entitat Autònoma del Diari Oficial i de Publicacions , the body that publishes Diari Oficial de la Generalitat de Catalunya (DOGC ) –the official journal of the Catalan government-, started its editing process adaptation to the electronic environment in 2009, based on Ley 2/2007 (DOGC, 5/6/2007), an act that defines the digital format as the only official format for DOGC. Two key projects in this process are presented: adoption and adaptation of EuroVoc thesaurus as an indexation tool, and the design of the new digital cycle of DOGC production and publication.

Bayesian network models for hierarchical text classification from a thesaurus

We propose a method which, given a document to be classified, automatically generates an ordered set of appropriate descriptors extracted from a thesaurus. The method creates a Bayesian network to model the thesaurus and uses probabilistic inference to select the set of descriptors having high posterior probability of being relevant given the available evidence (the document to be classified). Our model can be used without having preclassified training documents, although it improves its performance as long as more training data become available. We have tested the classification model using a document dataset containing parliamentary resolutions from the regional Parliament of Andalucía at Spain, which were manually indexed from the Eurovoc thesaurus, also carrying out an experimental comparison with other standard text classifiers.

Computer-aided Document Indexing System

An enormous number of documents is being produced that have to be stored, searched and accessed. Document indexing represents an efficient way to tackle this problem. Contributing to the document indexing process, we developed the Computer-Aided Document Indexing System (CADIS) that applies controlled vocabulary keywords from the EUROVOC thesaurus. The main contribution of this paper is the introduction of the special CADIS internal data structure that copes with the morphological complexity of the Croatian language. CADIS internal data structure ensures efficient statistical analysis of input documents and quick visual feedback generation that helps indexing documents more quickly, accurately and uniformly than by manual indexing.