This service provides detailed linguistic information for a given text in English, Spanish, French, Italian, Portuguese and Catalan. There are three operating modes that cover different aspects of the morphosyntactic and semantic analysis: Lemmatization, which provides the lemmas of the different words in a text; PoS tagging: which provides not only the grammatical category of a word, including semantic information about that word; Syntactic analysis: that provides a thorough syntactic analysis, giving a complete syntactic tree where the leaves represent the most basic elements and their morphological and semantic analyses.
Automatic multilingual text classification according to pre-established categories defined in a model. The algorithm used combines statistic classification with rule-based filtering, which allows to obtain a high degree of precision for very different environments. Three models available: IPTC (International Press Telecommunications Council standard), EuroVocs and Corporate Reputation model. Languages covered are Spanish, English, French, Italian, Portuguese and Catalan.
Topics Extraction tags locations, people, companies, dates and many other elements appearing in a text written in Spanish, English, French, Italian, Portuguese or Catalan. This detection process is carried out by combining a number of complex natural language processing techniques that allow to obtain morphological, syntactic and semantic analyses of a text and use them to identify different types of significant elements.
Multilingual sentiment analysis of texts from different sources (blogs, social networks,...). Besides polarity at sentence and global level, Sentiment Analysis uses advanced natural language processing techniques to also detect the polarity associated to both entities and concepts in the text. Sentiment Analysis also gives the user the possibility of detecting the polarity of user-defined entities and concepts, making the service a flexible tool applicable to any kind of scenario. Additionally, Sentiment Analysis detects if the text processed is subjective or objective and if it contains irony marks [beta], both at global and sentence level, giving the user additional information about the reliability of the polarity obtained from the sentiment analysis.