IntuView's core product, IntuScan is a decision-support expert system for real-time exploitation of documents in Arabic and other languages. Instantly assesses any Arabic-language document, determines whether it contains content of a terrorist nature or of intelligence value, provides a first-tier Intelligence Analysis Report of the main requirement-relevant elements in the document.
Tens of millions of non-English documents are seized each year in modern warfare and specifically in the war on terrorism. The challenge currently faced by Western intelligence agencies, militaries and law enforcement officers is the inability to rapidly extract and exploit information of high intelligence value contained in such captured documentation ("Document Exploitation" or DOCEX), whilst still in the field.
The current implementations of DOCEX is time consuming and suffers from several major deficiencies: long processing times, high percentage of false negatives and false positives, loss of intelligence due to ignorance of central cultural nuances on the part of the translators or analysts.
The IntuView Solution IntuScan, is an integrated multi-engine system. It is based on: a sophisticated Arabic language NLP engine tailored to the idiosyncratic neo-classical Arabic used by Islamic terrorists; a constantly updated domain-specific (of the field of Islamic terrorism) ontology and lexicon of entities, events, expressions and hermeneutics of religious texts; an extensive rule-base for disambiguation of lexical instances; continuous statistical modeling of an extensive and eclectic training set of Islamic documents and external data bases
The purpose of IntuScan is not mere "categorization" but generation of an intelligence summary of documents or batches of documents.
This summary includes:
Analyses of the significance of religious terms with terrorist ramification.
Phone numbers addresses and place references.
Indirect links of the document to elements not mentioned in the document.
Characterization and prioritization of the document.