The paper presents an ontology-based approach to monitor and assess changes and trends of technology intensive domains. The research, produced in collaboration with a variety of chemical engineering users of knowledge, promotes ontologies and semantically enriched models to enable knowledge sharing, classification, monitoring and reuse. Ontologies are deployed to search, retrieve and analyze technical structured and unstructured context from heterogeneous sources of information. The environment integrates models, language processing tools, knowledge engineering tools (crawlers, tokenizers), and agents. Besides the search and the management of information, the approach can be used to validate the suitability of ontologies and their relevance to the particular application, suggesting modifications and alterations when appropriate.

A generic version of the problem addressed with this paper could be stated as follows. Given is, (a) an ontology (OM) as a knowledge model that may account for the description of a project, system, task or assignment; the model accounts for the problem to tackle, and (b) a selected domain (D) that accounts for all the resources that contain potentially useful information related to this problem. The task would be then to (i) identify all subsets from (D) that relate to the problem, (ii) associate concepts and develop relationships between the components of OM and the information resources; and (iii) assess (OM) in relation to (D) and propose posssible upgrades that better match the domain.

For this purpose, the paper proposes a systematic and generic procedure that makes use of: (i) a reference domain, (ii) a knowledge model (ontology) of unknown and unspecified associations with the domain (iii) agents to process ontologies and assume tasks to collect, classify and cluster domain resources, (iv) Natural Language Processing tools that tokenize and annotate documents in preparation of stage analysis, (v) Analysis tools that process the results of stages (iii) and (iv). The procedure enables to map ontology concepts with the domains, review trends and relationships and present a list of possible modifications to the ontology. Developments are presented and integrated in the knowledge management platform (h-TechSight) that contains tools for editing ontologies. The figure above describes the general frame of developments and the applications presented. The implementation has integrated technologies in ontology editors, agent systems, annotation tools, and databases.  The work presents an ontology-based approach to manage and monitor chemical engineering knowledge. Conventional engineering concepts, such as flowsheets, are used to define and prototype ontologies used to search and manage information. The work takes up applications in general (advanced search engines) and targeted fields (technology watch, employment trends, monitoring tools). Preliminary results are highlighted in the paper. The concept of dynamic ontologies is new and is supported by search engines, and semantic web and reasoning tools available (WebQL, GATE IE, ad-hoc tokenizers, and crawlers) as developed in the course of the h-TechSight project.