This documentation supports the 20.08 version of BMC Helix Platform. 
To view an earlier version, select 20.02 from the Product version menu.

Search capabilities for custom applications

BMC Helix Platform provides the following search capabilities in your application:

  • Full-text search (FTS)—an index-driven search. When a user performs a search, only the index is searched instead of the entire text.
  • Global search—An application-level search that enables the user to perform search operations within an application.
  • Cognitive search—An AI-driven search that enables search by using natural language query, understands user's query, analyses data, and handles data enrichment requests for data change events.

Full-text search

Full-text search is a keyword-based search and involves indexing before searching so that when a user performs a search, only the index is searched instead of the entire text.  FTS can identify stem words and can return the results accordingly. For example, when you search for fire, the search results will include words such as firewall, backfire, firefighter, firecracker, and so on. FTS can weigh words according to their relevancy depending on where they appear—in a title, a keyword, or the text that is being searched. Results that match the title can be weighted as more relevant and can be displayed among the top search results. For FTS to search across certain fields, the fields must be FTS enabled.

Global search 

Global search is an application-level search that searches across multiple forms for records that match a word or phrase that a user types in the search area. Global search uses FTS and searches all fields that are FTS enabled in an application

Cognitive search capabilities

Cognitive search is an AI-driven search capability that utilizes artificial intelligence (AI) to merge the scattered knowledge and provide more relevant search results. You can leverage a cognitive search service in BMC Helix Chatbot and custom applications to enable cognitive search across BMC Helix Platform data and external data. 

Comparison between search capabilities

The following table provides the broad differences between FTS, global search, and cognitive search:

Full-text searchGlobal searchCognitive search
Full-text search is a keyword-based search.Global search is an application-level search that uses FTS to search for keywords. Global search searches across multiple record definitions.Cognitive search is a natural language search that provides enhanced results based on keyword extractions, sentiment analysis, relations, and so on.
You must enable fields for FTS so that the data in those fields is searchable during FTS.You must enable global search for searching terms in an application.

You must create collections in the cognitive search provider. Along with enabling the fields for cognitive search, cognitive search also requires search data sets that include searchable fields and record definitions.

Supports attachments in the following formats:

  • Adobe PDF
  • Microsoft Word
  • HTML
  • XML
  • OpenDocument
  • Electronic publication format (digital books)
  • RTF
  • Compression and packaging formation (.zip, .bzip2, .tar, .ar, .cpio)
  • Unicode
  • Java class files and archives (extracts class names and method signatures from Java class files and the .jar files containing them)
  • mbox (extracts email messages from the mbox format used by many email archives and UNIX mailboxes)
  • Image formats (extracts metadata from image formats supported by the Java platform)
  • Audio formats (extracts the lyrics, if present, and any metadata from MP3, MIDI, and other simple audio formats)
  • Video formats (supports only Flash video format using a simple parsing algorithm)

Supports attachments in the following formats:

  • Adobe PDF
  • Microsoft Word
  • HTML
  • XML
  • OpenDocument
  • Electronic publication format (digital books)
  • RTF
  • Compression and packaging formation (.zip, .bzip2, .tar, .ar, .cpio)
  • Unicode
  • Java class files and archives (extracts class names and method signatures from Java class files and the .jar files containing them)
  • mbox (extracts email messages from the mbox format used by many email archives and UNIX mailboxes)
  • Image formats (extracts metadata from image formats supported by the Java platform)
  • Audio formats (extracts the lyrics, if present, and any metadata from MP3, MIDI, and other simple audio formats)
  • Video formats (supports only Flash video format using a simple parsing algorithm)

Supports attachments in the following formats:

  • Adobe PDF
  • Microsoft Word
  • HTML

Related topics

Enabling full-text search capabilities in your application

Enabling global search for use in an application

Was this page helpful? Yes No Submitting... Thank you

Comments