Although text-based ontology engineering gained much popularity in the last 10 years, very few ontology engineering platforms exploit the full potential of the connection between texts and ontologies. We propose DAFOE, a new platform for building ontologies with a terminological component using di erent types of linguistic entries (text corpora, results of natural language processing tools, terminologies or thesauri). DAFOE supports knowledge structuring and conceptual modelling from these linguistic entries as well as ontology formalization. DAFOE outputs models with two main original features: an ontology articulated with a lexical component and a connection with the text or linguistic entry that motivated their de nition.
Design
DAFOE1 is a new platform for building ontologies using di erent types of linguistic entries (text corpora, results of natural language processing tools, terminologies or thesauri). DAFOE supports knowledge structuring and conceptual modelling from these linguistic entries as well as ontology formalization. DAFOE outputs models with two main original features: an ontology articulated with a lexical component and a connection with the text or linguistic entry that motivated their de nition. The requirements of the platfom and its development focus 1) on integrating various kinds of tools currently used within a single modelling platform, 2) on guaranteeing persistence and traceability of the whole ontology building process, and 3) on developing the platform in an open source paradigm with possible plugin extensions. 2.
TEXT-BASED ONTOLOGY ENGINEERING
There is a growing interest for ontologies and related tools, including Ontology Engineering Environments. Many of them are pure ontology editors that support the development of formal ontologies but do not assist the tasks of knowledge acquisition or structuring. Knowledge engineers are supposed to have a rst of ontology draft before using such tools. Since 2003, a signi cant shift occurred. Firstly, a parallel has been established between ontology population from text and semantic (textual) annotation. Secondly, many projects have proved the bene t brought by Human Language Technologies (HLT), including NLP, Information Extraction, Knowledge Discovery or Text Miming, for complementarity activities such as ontology learning from text and ontology population. The diversity and richness of existing HLT tools as well as the complexity of the ontology development tasks underlined the need for tool suites and platforms where the knowledge engineer can de ne its modelling strategy.This challenge is also one major motivation of the DAFOE project but the platform targets more ambitious goals: a better interoperability, a higher robustness and an easier combination of HLT and ontology technologies.
The goal of DAFOE is both to extend the variety of HLT that can be used and to support scalable ontology engineering. It claims that there are several ways to get an ontology, and that tools and processes must be selected according to each ontology case-study. DAFOE will propose tools similar to those of Text2Onto, but human supervision will play a major role for selecting tools, validating their results and conceptualizing. Knowledge conceptualization requires that a human selects and organizes properly concepts and relations, but this process can be guided. The result of DAFOE will typically be a termino-ontological resource where the ontology is connected to a lexical component. 3.
DAFOE data model has to take into account various ontology building strategies, whatever information source (texts, terminologies, thesauri or human expertise) is used. 3.1
The data model is based on a valid methodology for
building ontologies from texts, which has inspired tools such as
Terminae [
Entry Entry Entry Entry
API for NLP tools
Imports Layer 0 (corpora) Text visualization Layer 1 (terminological) Linguistic study Layer 2 (terminolo-conceptual) Semantic network building SKOS export Layer 3 (ontological) Formal ontology building
OWL export 3.2
The corpora layer is useful for the knowledge engineer willing to build an ontology from text. He/she can build a working corpus by selecting di erent source documents and browse that corpus, either as plain documents or as segmented ones. In the data model the corpus is represented as a sequence of sentences, each one having a unique identi er. 3.3
The terminological layer gives a view over the domain
speci c lexicon of the corpus. It gathers the terms of the
domain and their relationships. Terminological knowledge is
traditionally produced by NLP tools such as term extractors
applied on the working corpus. The underlying assumption
is threefold: text analysis can extract term candidates that
are relevant for a given domain, those terms are likely to be
turned into ontology concepts and the distribution of these
terms re ects their semantics [
This layer represents a semantic structure of unambiguous termino-concepts (TC) and termino-conceptual relations (RTC). The knowledge engineer may build that layer by importing a preexisting termino-conceptual resource such as a thesaurus or out of the analysis of the terminological layer. In that case, he/she analyses the meaning of terms and relations that appear at the terminological layer with respect to each other and by looking at their occurrences. The termino-conceptual layer is pivotal for transforming linguistic elements into conceptual ones and tracing the ontology back to the linguistics. 3.5
The ontology data model allows to formalize TCs and RTCs in a formal language equivalent at OWL-DL. Concepts are described as classes, individuals as instances of classes, properties between classes as object properties and properties between a class and a value as data properties or attributes. An automatic process will translate TCs and RTCs into formal concepts in a hierarchy with inherited properties as usual subsumption in description language. This translation exploits the structure of the semantic network represented in the termino-conteptual layer and the di erential criteria associated with TCs and RTCs. 4.
A prototype of the DAFOE platform has been implemented. DAFOE is intended to provide a variety of ontology engineering methods. As such a diversity can not be managed in a unique and static model, we adopted an extended Ontology-Based Database (OntoDB) architecture that supports model management and plugins. The strength of DAFOE approach is i) a precise de nition of the various steps by which one can design a formal ontology; ii) a data model guaranteeing persistence and traceability of the whole ontologie building process; iii) the supply of exible methodological guidelines that support the knowledge engineer without constraint; iv) an architecture based on the MOF model and plugins adaptability to ensure extensibility of the model and processes around a core tool; v) the speci cation of various modelling strategies based on di erent input/output of the platform; vi) the nal production of an ontology which is associated to a terminological component. 5.