Entities

Content item

• Object which represents the content to be enhanced by Apache Stanbol

Content parts

• Are used to represent the original content as well as transformations of the orig. content

Analysed text

• Used as content part
• Describes
  • Structure of the text, such as text-sections, sentencens, chunks and tokens
  • Annotations for the detected text

Enhancer

• Allows to extract features from passed content
• Uses enhancement engines based on the called enhancement chain

Enhancement engines

Preprocessing

• Content type detection
• Text extraction

Natural Language Processing (NLP)

• Language detection

  • Adds language annotation (as defined by STANBOL-613) to the metadata of a content-item

• Sentence detection

  • Adds sentences to the analyed text content part

• Tokenizer engines

  • Adds tokens to the analyzed text content part

• Part of speech tagging

  • Zuordnung von Wörtern und Satzzeichen zu Wortarten (Verb, Substantiv etc.)

• Chunk / prhase detection

  • Adds detected chunks to analyzed text content part
  • Annotate added chunks with the type of the detected phrase

• Named entity recognition (NER)

  • Writes detected named entities as annotations to the metadata of the content-item

• Morphological analysis

  • Performes lemmatization (lexikographische Reduktion der Flexionsformen eines Wortes auf eine Grundform)

• General NLP processing engines

Linking / suggestions

• Suggestion of entities for features present in the parsed content

• Provides

  • Type
  • Description
  • Spatial and/or temporal content
  • Links to other entities

• Named Entity

  • Suggests links to several linked data sources

• Entityhub

  • Suggests links to entities managed by entity hub, referenced sites or managed sites

• FST (Finite State Transfer)

  • Links entities indexed in a Solr database

• Entity co-mention

  • Detects co-mentions of a deteceted entity at a later position

• DBPedia spotlight annotation engine

  • Includes NLP, Entity Linking and Disambiguation of Entities using DBpedia as knowledge base

• Geonames

  • Suggests links to geonames.org

• OpenCalais

  • Integerates services from Open Calais
  • Provides both NER and entity linking

• Zemanta

  • Integrates services from Zemanta
  • Provides both NER and entity linking

• Sentiment analyses

  • Engines that perform word/chunk level sentiment classifications on the analyed text

• Disambiguation

  • Disambiguation (Auflösen von Mehrdeutigkeiten) entities based on contextual information

Postprocessing / other

• Not yet covered :)

Enhancement structure

• Represents the state of enhancement based upon the content item and the content parts
• Gets constantly updated during the enhancement process

General information

Used namespaces

• fise

  • This is the main namespace of the currently used Enhancement Structure. All custom concepts and properties are defined using this namespace.

• enhancer

  • This is the main namespace of the Stanbol Enhancer defining concepts such as ContentItem, EnhancementEngine, EnhancementChain …

• entityhub

  • This is the main namespace of the Stanbol Entityhub component.

• dc

  • The Dublin Core terms standard is also heavily used by the Stanbol Enhancement Structure. Especially to encode metada data, but also to encode relations between extracted information

• dppedia-ont

  • Concepts of this Ontology are used to describe the types of "Named Entities" detected in parsed content.

• skos

  • The SKOS standard is preferable used to describe entries of Thesauri or more generally any type of controlled vocabularies.

• rdf