Abstract  We propose new methods to detect paradigmatic fields through simple statistics over a scientific content database. We proposean asymmetric paradigmatic proximity metric between terms which provide insight into hierarchical structure of scientific activity and test our methods on a case studywith a database made of several millions of resources. We also propose overlapping categorization to describe paradigmaticfields as sets of terms that may have several different usages. Terms can also be dynamically clustered providing a high-leveldescription of the evolution of the paradigmatic fields.

We show that the singular value decomposition of a term similarity matrix induces a term taxonomy. This decomposition, used in Latent Semantic Analysis and Principal Component Analysis for text, aims at identifying “concepts�? that can be used in place of the terms appearing in the documents. Unlike terms, concepts are by construction uncorrelated and hence are less sensitive to the particular vocabulary used in documents. In this work, we explore the relation between terms and concepts and show that for each term there exists a latent subspace dimension for which the term coincides with a concept. By varying the number of dimensions, terms similar but more specific than the concept can be identified, leading to a term taxonomy.

We present a novel approach to the automatic acquisition of taxonomic relations. The main difference to earlier approaches is that we do not only consider one single source of evidence, i.e. a specific algorithm or approach, but examine the possibility of learning taxonomic relations by considering various and heterogeneous forms of evidence. In particular, we derive these different evidences by using well-known NLP techniques and resources and combine them via two simple strategies. Our approach shows very promising results compared to other results from the literature. The main aim of the work presented in this paper is (i) to gain insight into the behaviour of different approaches to learn taxonomic relations, (ii) to provide a first step towards combining these different approaches, and (iii) to establish a baseline for further research.

The application of clustering methods for automatic taxonomy construction from text requires knowledge about the tradeoff between, (i), their effectiveness (quality of result), (ii), efficiency (run-time behaviour), and, (iii), traceability of the taxonomy construction by the ontology engineer. In this line, we present an original conceptual clustering method based on Formal Concept Analysis for automatic taxonomy construction and compare it with hierarchical agglomerative clustering and hierarchical divisive clustering.