Wizards NLP Content Classification



A Multi-class Text Classification Approach

Wizards Content Classification provides context-sensitive analysis and automation features to organize unstructured content. The solution categorizes unstructured information by analyzing the full text of documents and Webpages and applying rules that automate classification decisions. It can organize information by policies or key words, and it can assign metadata that is based on, for example, the full context of a document. The solution works with Wizards Sell-side Platform to gather training data.

In this demo, we are excited to be sharing our hard-earned insights on building an industry leading classifier. We show you how we trained a model to make educated guesses of higher probability thresholds and classify miscellaneous articles into their respective categories- automatically, in a fraction of the time it would take to perform this manually, and error-free.

Paste any document or article below


Semantic Data Modeling

Wizards Content Classifier categorizes unstructured information by performing Knowledge graph-powered semantic analysis over the full text of the documents and applying supervised machine learning and rules that automate classification decisions.

Knowledge and Awareness

Represented as a knowledge graph that contains factual information: diverse data about rich set of entities and concepts of interest. Possessing enough lexical and grammatical knowledge to comprehend text, combined with the skill to resolve ambiguity and extract relationships from free text.

Content Classification Relationship Graphs

Classify documents by common entities or 55+ general categories such as News, Technology and Entertainment. Build relationship graphs of entities extracted from news or wikipedia articles, by using signals from the state of the art syntax analysis.


Linking Text and Knowledge Graphs to Deliver Actionable Insights

Semantic Search, Exploration, Categorization and Recommendation

Wizards Content Classification Platform Conceptual Architecture

  • Wizards Content Classification Platform consists of a set of databases, machine learning algorithms, APIs and tools we use to build various solutions for specific enterprise needs.
  • Fluidtable as a semantic graph database for storing semantic indexes, consolidated entity profiles, linked open data and user behavior profiles;
  • Machine learning models for text analysis, disambiguation of entities, concept extraction and classification;
  • APIs for text analysis, model training, search, recommendations, content management and concept profiles;
  • Tools used to build UIs for contextual authoring, enrichment monitoring, curation and quality assurance, template definition and other user applications;
  • We are able to ingest and normalize content and data from any number of diverse sources.
  • We create user profiles based on user activity and combine them with contextual similarity derived from the semantic analysis.
  • Data Driven Targeted Advertising and Recommender Engine (ATARE).

    Copyright © 2016 - 2019. Cubex Ltd.