SAS^® Enterprise Content Categorization

Features

Entity, fact and event extraction

• Locate and return related pieces of data that form a fact or an event based on their context in real time (e.g., a person in relation to a company, the merger of two companies, etc.).
• Identify unknown information without a requirement for precompiled dictionaries.
• Customize matching criteria using options such as contextual markers, parts of speech tags and Boolean operators.
• Use prebuilt co-reference operators in an intuitive GUI to help resolve pronouns more easily, addressing syntactical functions such as:
  • Linking to a matched string with its canonical form.
  • Co-referencing classifier definitions.
  • Restricting forward and preceding co-reference matches.
  • Assigning a new concept name for a match on a specific term.
• XML fields can be limited for specified matches.
• Case-insensitive operators allow for greater rule-matching precision, including:
  • Prebuilt stemming can match on all word forms or only on noun or verb forms.
  • Sentence and paragraph operator defines the number of word, noun or verb matches within a paragraph.
• Write more than one rule to extract all of the possible permutations of the data you seek.
• Disambiguate facts and events by excluding certain matches.

Category classification

• Control the number of terms that an automatic category rule generates for each category.
• Adjust relevance schemes (frequency-based or zone-based) for concepts used in category rules.
• Use linguistic rules, which are unique identifying terms, or add Boolean operators to your unique terms for added specificity in determining category membership.
• Develop a list of unique identifying terms for each category rule.
• Weight selective terms or the categories themselves, creating more exclusive membership requirements.
• Use the testing facilities to validate application of rules and definitions to batch, entire or content components.
• Includes enhanced XML handling along with better syntax checking and duplicate rule elimination for classifier concepts.
• Automatically apply the rules and definitions to incoming texts using the client APIs in C, C++, C#, .NET, Java, Perl or Python.

Collaborative and flexible taxonomy management

• Define user-permission levels, including read, write, category rules and concept definitions.
• Use an unlimited number of taxonomy nodes, and apply categories and concepts generated to large volumes of input documents.
• Develop a hierarchical taxonomy where related topics are grouped together, or create a flat taxonomy where there is no relationship between any of the nodes in the taxonomy tree.
• Automatically generate categories from Wikipedia data to jump-start your taxonomy development.
• Benefit from improved PDF conversion.

Out-of-the-box integration

• Configure APIs to automatically tag content from Microsoft Office SharePoint, Endeca, FAST ESP, Documentum or other systems.
• Documents are tagged prior to indexing to speed up processing time.
• Extends capabilities of existing search tools and content management systems, increasing the relevance of retrieved materials.

Support for more than 27 languages

• Language tools: NLP/advanced linguistic technologies that leverage:
  • Part-of-speech recognition and tagging: Recognizes nouns, verbs, adjectives, etc.
  • Stemming: Locates the various forms of an input noun or verb.
  • Case sensitivity: Specifies uppercase and/or lowercase recognition for concepts.
• Use the following two options with Germanic and Asian languages:
  • Compound recognition and compound decomposition: Break apart the recognized compound words.
  • Segmentation for Asian languages.
• The product ships with English and the native language if other than English. Additional languages are licensed as add-ons.

Add-on modules available

• Multiple, prebuilt industry-specific taxonomy starter kits are available as add-on modules for SAS Enterprise Content Categorization:
  • Provide immediate ROI by classifying industry-specific content to help jump-start document classification initiatives.
  • Include detailed concepts and attribute values with predefined rules, helpful for initiating taxonomy project development.
• Search and indexing add-ons automatically discern query semantics and enable superior drill-down and investigative capabilities by categorizing multifaceted information:
  • Include an easy-to-use interface for search and document processing.
  • Can extract entities, concepts and facts within a unified document processor.
  • Support multiple document schemas with multiple instantiations, and also retain the original URL in any split documents.
• Web crawling add-ons automatically download documents from the Internet by performing several different kinds of downloading operations based on network bandwidth, crawling politeness and information coverage:
  • Have an easy-to-use interface for defining and managing Web and internal file system crawls.
  • Set individual quotas for a specified URL, and define project quotas.
• Other add-on modules include text summarization, document duplication detection, content alerts and additional languages.