TEMIS Powers Thomson Scientific ISI Web of KnowledgeSM indexing

TEMIS, a leading provider of Text Analytics applications and Thomson Scientific, part of The Thomson Corporation and leading provider of information solutions to the worldwide research and business communities, today announced that they have entered a software licence and services agreement.

Thomson Scientific has chosen TEMIS acclaimed indexing technology to power its content processing system.

Thomson Scientific provides access to high-value, essential information for researchers and scholars worldwide through integrated information solutions delivered by the most innovative technologies. Thomson Scientific’s ISI Web of KnowledgeSM is the multidisciplinary single environment from which researchers can access, analyze, and manage information.

Because Thomson Scientific required a robust Indexing solution that could easily be integrated into their custom-made workflow, they turned to TEMIS to automate its data sources processing. The solution had to leverage a dictionary of over 2 million items, and annotate more than 100.000 documents per week, in order to finally produce indexed, visible and searchable content for its ISI Web of KnowledgeSM portal.

“At Thomson Scientific we are always exploring new innovative ways of enriching our content. As a market leader in analytical tools space, we chose TEMIS because it met our strict precision as well as throughput requirements. Furthermore, the flexibility of the TEMIS Text Analytics framework enables us to continue to evolve the platform as our content and needs change”, said Sina Adibi, CTO and SVP of Systems and Technology of Thomson Scientific.

Meeting Thomson Scientific’s requirements in term of quality, scalability and robustness, TEMIS’s Insight Discoverer™ Extractor and Insight Discoverer™ Categorizer were held as the most appropriate Indexing solution, providing pre-packaged annotators (Skill Cartridges™), and the ability to easily add lexicons.

Insight Discoverer™ Extractor and Insight Discoverer™ Categorizer are highly scalable servers. The extraction server processes documents* of any format to annotate and enrich them with metadata, such as entities, relations, categories, topics, and attributes. The categorization server automatically assigns pre-defined categories to documents according to their semantic profile. Finally, TEMIS Text Analytics solution leverages Mondeca’s Intelligent Topic Manager™, a powerful ontology management solution, to effectively store, and edit terminologies and taxonomies.

“TEMIS is proud to be part of this highly innovative content processing project for ISI Web of KnowledgeSM.“, said Guillaume Mazières, VP Sales and Marketing at TEMIS. “Our Text Analytics solutions help Thomson Scientific provide comprehensive and actionable knowledge that not only support but also accelerate scientific Discovery.”

TEMIS’s Text Analytics solution was deployed to index the BIOSIS Digital Archive, including over 2 million scientific documents, according to a wide range of terms types such as organisms, chemicals, diseases, geographical locations, etc. It was used in conjunction with a set of custom and standard Skill Cartridges™ (Medical Entity Relationships, Biological Entity Relationships, Text Mining 360°) to enrich BIOSIS content while updating the legacy metadata.

Thomson Scientific is now considering other enhancements to its editorial processing to support its team of indexers with a pre-step to manual indexing.

* Journal articles, abstracts, patents, news items, web content, etc.