The agreement stipulates that Convera will embed its XeLDA linguistic engine in four languages (French, Spanish, Russian and German) into its Excalibur Web Search Platform.
XeLDA is a multilingual linguistic engine. It models and standardizes unstructured documents in order to automatically exploit their content. Based on a technology developed through 20 years of research and development, it provides advanced Text Mining features enabling textual information processing.
XeLDA offers a scalable range of services based on natural language processing components that can be integrated in business applications:
* Automatic identification of the language within each document
* Segmentation of text into sentences
* Split of text into basic lexical units (tokenization)
* Morphological text analysis to return the normalized form (the lemma) and the potential grammatical categories for all the words identified during the tokenization stage
* Morpho-syntactic disambiguation to determine the exact grammatical category of a word according to its context
* Extraction of sequences of words that form noun phrases
* Identification of the context of a word to find the corresponding dictionary entry (Dictionary lookup)
* Recognition of idiomatic expressions
“Even though English remains the lingua franca for the Web, it is increasingly becoming multilingual in nature, “said Claude Vogel, Chief Technology Officer at Convera. “We are pleased to be working with Temis to meet linguistic requirements for authoritative search products.”
Temis technology provides superior results, using its award-winning and patent protected linguistic technology as well its packaged Skill Cartridges for domain-specific analysis. Temis linguistic technology is today available in 20 languages, including Chinese, Japanese, Korean and Arabic.
www.Temis.com
