INTELARTES


Innovation  &  Technology in Semiotics
Technology


Based on its extensive R&D experience in intelligent systems, Intelartes develops an integrated approach to knowledge-based information processing. The approach is an analogy of engines, fuels and refinery. The engine, a knowledge-based system, is intended for applications of complex operational logics. It is fuelled by knowledge models of generic or specific operational logics. The knowledge, the fuel of the engine, is engineered in the refinery, where a set of processes, methods and tools are used in knowledge engineering.

Engines

The engine is a knowledge based system. It processes data and information intelligently, with reference to the knowledge model deployed in the system.

Intelartes Engine
Functionality
Parts
Text

Text engine transforms files of various formats into pure texts, segments them and stores in them in text database for further analysis and processing, such as, building content profiles of each segments, indexing or grouping texts by a given set of thematic tags.

  • Transformer
  • Segmentor
  • Profiler
  • Classifier
  • Text store
Language The language engine tokenizes a set of sentences, tags the lexical tokens by parts of speech, parses to build phrase structures and functional relations, and interpret the lexical functional features and relations into semantic structures.
  • Tokeniser
  • Tagger
  • Parser
  • Interpreter
Semantics Semantics engine interprets inputs (parses, structured data) into a network of conceptual entities and relations in terms of an application specific semantic model or generic ontology. It computes the semantic similarity and difference of the inputs, combines and matches semantically.
  • Comparator
  • Matcher
  • Translator
  • Semantic Base
Web Utilities Web processing facilities features capabilities to search over different search engines and crawls the relevant pages for contents collection. The crawler makes use of resources available to language and text engines.
  • Meta Search
  • Crawler

Knowledge

The fuel for the knowledge-based engine is the knowledge base, modeled for it for a specific application purpose, from the knowledge resources of the application domain.

Knowledge bases

Intelartes engines require two types of knowledge bases. One is about languages; the other is domain/application semantics.

 

Fuel
Repository
Purpose
Type
Linguistics

Morphology

Morphology for processing the inflexion and conjugation of derived forms of words

  • French
  • English *
  • Chinese *
  • Dutch *
  Lexicon Lexicon for tokenization and tagging of lexical tokens
  Grammar Grammar for structural and functional parsing
  Synonym Synonym for computing semantic affinity
Semantics Ontology Ontology as domain framework for modeling application semantics and as basis for semantic operations
  • Activity Semantics **
  Application semantics Application semantics as model or meta data schema about semantic entities and relations to tag, compare and interpret data
  • Frameworks of occupation and competences **
Translation Multilingual Dictionary Dictionary for translating semantics in the source language into the target language  
  Case base of translations Case-based reasoning for semantic translation.  
    * under development ** continued work