Identify and extract key concepts from content

Nconcept extractor semantically adds layers of meaning to content

Overview

Nconcept extractor is a linguistic and statistical analysis engine, used to identify the central concepts (uncontrolled vocabulary) contained in any document. It achieves this by separating words and delimiters, identifying parts of speech (nouns, verbs etc.) and parsing complete, complex phrases according to linguistic patterns.

Applications

Concept clustering: By extracting concepts and creating rich metadata with them, concepts can be clustered into meaningful hierarchical groupings. This enables the quick classification of content and promotes increased relevancy across documents.

Contextual Advertising: Using Nstein's Nconcept extractor, Ncategorizer and Nfinder, publishers no longer have to manually specify relevant keywords for each asset that are linked to ads. Instead, relevant concepts, topics and/or entities are extracted and then matched to advertising. This translates into higher CPM and increased click-through.

Search Engine Optimization: Descriptive metadata is central to SEO ranking. Strategically placed keywords and phrases can significantly boost SEO rankings in search engine results. Drastically improve SEO ranking by providing consistent and descriptive metadata that many content providers lack.

Syndication: By appending relevant concepts and categories to assets, Nconcept extractor increases syndication opportunities. Organized, classified content is easier to sell and syndicate to aggregators.

How Nconcept extractor works

Every document contains particular words or phrases that capture its meaning. Because of its ability to understand linguistic context, Nconcept extractor is able to isolate these words.

First, a document is parsed to identify words and delimiters.

It is further segmented, this time different parts of speech (nouns, verbs etc.) are identified.

It then identifies simple or complex concepts, based on word or phrase context according to linguistic patterns.

Linguistically extracted concepts are subsequently weighted and ranked by intelligent algorithms tuned to identify relevancy based on frequency, quality of filtered concepts, distribution and more.

Input/Output

Nconcept Extractor ingests documents in any format, and outputs:

• A list of keywords and key phrases with relevancy rankings