Cognitive Search in your apps

Version 1
    Share This:

    The Cognitive Search Service is now included with BMC Helix Platform for use in custom applications. You can perform a cognitive search across BMC Helix Platform data and external data (file system and database based) by using natural language query. Cognitive search understands your query and analyzes data. It also handles data enrichment requests for data change events.

     

    Cognitive search capabilities:

    • Crawl, convert, enrich, and normalize data.
    • Customize how your documents are indexed to better target relevant information.
    • Securely explore your proprietary content.
    • Apply the following additional enrichments to the data published to IBM Watson Discovery:
      • Keyword extraction
      • Entity extraction
      • Concept tagging
      • Sentiment analysis through Natural Language Understanding (NLU)
      • Relations
      • Category classification

     

    Enabling cognitive search

     

    As an administrator or application business analyst, you can configure cognitive search as an alternative to full-text search (FTS). Cognitive search merges data in different locations and formats, and also normalizes and enriches the data. Your application users are more likely to find answers to questions because more of your scattered organizational information is made available to them. This enables your application to search in more than one data set.

    BMC Helix Platform cognitive service provides support for external and internal data sets in the cognitive search configuration of your application. A data set is a collection of documents. BMC Helix Platform supports the following types of data sets:

    • External data sets: An external data set is information created outside of BMC Helix Platform. You can link multiple external data sets from multiple sources to your application. For example, you can create a data set of your Remedy knowledge articles; you can create a data set from the content of one or more websites; you can also create a data set that contain PDF, Word, PowerPoint, Excel, PNG, TIFF, JPG, JSON, and HTML content.
    • Internal data sets: An internal data set is the record instance data from BMC Helix Platform. You can create your own collections of knowledge articles specific to your business needs and include the internal data sets in the cognitive search configuration of a chatbot application.

     

    You must perform the following tasks to enable cognitive search in your BMC Helix Platform based applications:

    TaskActionReference
    1As an administrator, configure the cognitive search service.Configuring a service provider for cognitive search
    2

    As an application business analyst or an administrator, enable cognitive search on the application's data defined by record definitions and indicate which fields are searchable.

    Enabling cognitive search in fields
    3

    As an application business analyst or an administrator, if you want to include external data (documents and database tables for external applications) in a cognitive search, then create an IBM Watson Discovery collection and use the IBM Watson data crawler to upload this data to the IBM Watson Discovery collection.

    Creating file system and database collections for searching external data
    4

    As an application business analyst, create an in-bundle setting in the application so that administrators can define the search data sets.

    The search data set is an abstraction of the IBM Watson Discovery collection.

    Creating an application configuration for defining search data sets

    5

    As an application business analyst, define search data sets for BMC Helix Platform data and external data.

    One search data set corresponds to one IBM Watson Discovery collection.

    Defining search data sets

     

    After you have completed the above setup, your application can use the cognitive search functionality.