“Language” is often a political and not a linguistic definition; the different languages - Swedish, Danish and Norwegian – are more inter-intelligible than Moroccan, Libyan, Mauritanian and Egyptian which are all called “Arabic”. In addition, languages are replete with cultural memes that are not evident to the non-initiated. These language “registers” can be rooted in social status, education, age, profession, and geography. Furthermore, even within the same language register, words, quotations, idioms or historical references can be “polysemic”; they have different meanings according to the domain and the context of the surrounding text.
IntuView’s core technology incorporates our own propriety Natural Language Processing (NLP) engines with domain-specific ontologies to perform real‐time semantic text analytics in a variety of supported languages. IntuView does not settle for “data mining” or extraction of word strings from documents but performs sophisticated “meaning mining” on the text in order to extract not merely the simple meaning but the underlying and implicit “hermeneutics” of the text.
The purpose of IntuScan (the NLP engine of IntuView) is to extract all relevant information from of the texts. performs deep categorization of the document according to various parameters (genre, topic, political/ideological leanings). This categorization (domain detection) provides information that enables the Natural Language Processing engine to achieve a higher level of accuracy in disambiguation of polysemy in the texts. It also allows the user to focus on the texts that really are important without the need to invent ambiguous "keywords" and crawling rules. IntuScan extracts, resolves, aggregates, disambiguates and collects implicit and contextual information about entities of all types (persons, organizations, places, events, ideas etc.), including relations between them. It also performs deep nuanced sentiment analysis (who thinks what about what characteristic of whom).a large number of unstructured texts. The output is a structured searchable representation as well as human-readable.
IntuView's technology is patented in the USA ( # 8078551 ) and Israel (#189786 ).
The results of the analysis of the texts are inserted into a database, thus turning the unstructured text into a structured format. This database can then be searched by the IntuView Search and Analysis Platform.
Download the file below for more information.