Skip to content

Latest commit

 

History

History
141 lines (102 loc) · 11.5 KB

getting-started.md

File metadata and controls

141 lines (102 loc) · 11.5 KB
Raw
copyright lastupdated subcollection
years
2019, 2021
2021-03-14
discovery-data

{:shortdesc: .shortdesc} {:external: target="_blank" .external} {:tip: .tip} {:note: .note} {:pre: .pre} {:important: .important} {:deprecated: .deprecated} {:codeblock: .codeblock} {:screen: .screen} {:download: .download} {:hide-dashboard: .hide-dashboard} {:apikey: data-credential-placeholder='apikey'} {:url: data-credential-placeholder='url'} {:curl: .ph data-hd-programlang='curl'} {:javascript: .ph data-hd-programlang='javascript'} {:java: .ph data-hd-programlang='java'} {:python: .ph data-hd-programlang='python'} {:ruby: .ph data-hd-programlang='ruby'} {:swift: .ph data-hd-programlang='swift'} {:go: .ph data-hd-programlang='go'}

Getting started with the Watson Discovery Sample Project

{: #getting-started}

In this short tutorial, we introduce the {{site.data.keyword.discoveryshort}} Sample Project. The Sample Project is a great way to tour and try out {{site.data.keyword.discoveryshort}} features. {: shortdesc}

This information applies to {{site.data.keyword.discovery-data_short}} and Premium instances that were created on {{site.data.keyword.cloud_notm}} after 16 July 2020. For Premium instances that were created before that date and for all Lite and Advanced plans, see Getting started with Discovery{: external}. {: important}

Before you begin

{: #before-you-begin-tool} {: hide-dashboard}

Choose the appropriate step to complete for your deployment:

Step 1: Open Watson Discovery

{: #getting-started-launch-tool}

Cloud Pak for Data only {{site.data.keyword.icp4dfull_notm}}

{{site.data.keyword.discoveryshort}} installed on {{site.data.keyword.icp4dfull}} 3.0.1 or later:

  1. From the {{site.data.keyword.icp4dfull_notm}} web client menu, choose My Instances.
  2. Select your {{site.data.keyword.discoveryshort}} instance.
  3. Click Launch tool.

{{site.data.keyword.discoveryshort}} installed on {{site.data.keyword.icp4dfull}} 2.5.0.0:

  1. From the {{site.data.keyword.icp4dfull_notm}} web client menu, choose My Instances.
  2. On the Provisioned instances tab, find your {{site.data.keyword.discoveryshort}} instance, and then hover over the last column to display the ellipsis icon Ellipsis icon and choose View details.
  3. Click Open Watson Discovery.

![IBM Cloud only](images/ibm-cloud.png) **{{site.data.keyword.cloud_notm}}**
  1. {: hide-dashboard} Click the {{site.data.keyword.discoveryshort}} instance you created to go to the service dashboard.
  2. {: hide-dashboard} On the Manage page, click Launch Watson Discovery. If you're prompted to log in to the tooling, provide your {{site.data.keyword.cloud_notm}} credentials.

Step 2: Open the Sample Project

{: #open-project-tool}

  1. Open the Projects page by selecting My Projects. For more about projects, see Creating projects.
  2. Select Sample Project. The Improve and Customize page opens.

If you just installed {{site.data.keyword.discoveryshort}}, the Sample Project will need time to finish processing documents. Wait until processing is complete to start experimenting. You can confirm that all 40 documents have finished processing by selecting the Manage collections icon on the navigation panel and checking the Sample Collection. {: note}

The Project type of the Sample Project is Document Retrieval. Document Retrieval projects are used to search and find the most relevant answers from your data. It includes one collection that contains 40 documents.

Document Retrieval projects have the following defaults (we'll learn more about these later):

  • Settings: Optical Character Recognition (OCR) off
  • Enrichments applied: Entities, Parts of speech
  • Improvement tools enabled: Facets (by Entity), Dynamic Facets, Passages

A Document Retrieval project can contain a maximum of 5 collections.

There are three other Project types: Conversational Search, Content Miner, and Custom. {: tip}

Step 3: Tour the Sample Project

{: #tour-project-tool}

Now that we've opened the Sample Project and learned more about Document Retrieval projects, let's explore.

  1. Select the Manage collections icon on the navigation panel. Choose the Sample Collection. Several tabs display but we'll concentrate on a few.
    • The Activity tab displays collection details: Number of documents, Collection status, Date of last update, an abbreviated list of Warnings and errors (to see the full list, select View all). For more information, see Collection activity.
    • The Identify fields tab is where you can annotate your document using Smart Document Understanding. You can use the Field labels that are available or create your own.
    • The Manage fields tab gives you the option to choose the fields you want to index, turn on document splitting, and set date formats. For more information, see Managing fields.
    • The Enrichments tab displays the available enrichments (you can create more). By default, the enrichments already applied to this collection include Entities and Parts of speech. Click on Fields to enrich and you will notice that these enrichments are applied to the text field. To learn more about these enrichments, see Extracting meaning. You can also create enrichments that will add related terms (Dictionary), identify and extract values (Regular expressions), extract entities and relationships/apply rules to fields in your collection (Machine Learning and Watson Explorer Content Analytics Studio models), classify your documents into categories (Classifier), or use an Advanced rule model. See Creating enrichments.
  2. Select the Integrate and Deploy icon on the navigation panel. From here, you can share your project with colleagues and deploy. For more information, see Building and deploying components.
    • The Share preview link tab makes it easy to share your project with others. Follow the instructions to add a user, then send those login credentials and the provided link to your colleague.
    • The Explore UI components tab includes several componets: Search bar, Search results, Facets, and Document preview. The code for the {{site.data.keyword.discoveryshort}} components can be found on GitHub{: external}. You can learn more about the components and preview them in Storybook{: external}.
  3. Select the Improve and customize icon on the navigation panel. The Improve and Customize page is where you can to try out queries, then add and test customizations to improve the query results for your project. For more information, see Customizing and improving your project.
    • There are several example queries in the Try an example to get started section. For Watson, click the Run search button.
    • Two of the default Improvement tools for this project (Facets (by Entity) and Dynamic Facets) have identified facets (IBM, Watson, Kubernetes) you can filter on. Select a few and try them out.
    • Query results display in the center pane. Another Improvement tool, passages, makes it possible to see the passages identified in the documents returned. Click View passages in document for the document preview. For more information about how passages are identified in natural language queries, see Passages.
    • On the Improvement tools panel, select the dropdowns to see what is available to further customize your project. There are five categories of tools: Customize display, Search results, Teach domain concepts, Define structure, and Improve relevance. For the full list of tools, see Improvement tools.

That's the tour, stay on the Improve and Customize page for the next step.

Step 4: Customize the Sample Project

{: #customize-project-tool}

Now, let's customize a bit. Starting on the Improve and Customize page ...

  1. Enter the natural language query How do I install {{site.data.keyword.icp4dfull_notm}} in the query box.
  2. Review the query results displayed. View the source document for each result by clicking on View passage in document.
  3. On the Improvement tools panel, choose Customize display, and select Facets. Select New facet, then click the From existing fields in a collection button. Choose enriched_text.entities.type from the dropdown and click Apply. The new facets will display. You can change the name of the facet label, and the filtering behavior.

Step 5: Keep exploring and experimenting

{: #next-steps-tool}

Continue experimenting with the Sample Project. A few other things you might want to try:

  • Select the Manage collections icon on the navigation panel. Click the New collection button. Choose Upload data and create a new collection with a few of your own documents.
  • Select the Integrate and Deploy icon on the navigation panel. On the Share preview link tab, copy the link and paste it in your browser to preview your application. On the Explore UI components tab, click the Try it on Storybook link for any of the components.
  • Select the Improve and customize icon on the navigation panel. On the Improvement tools panel, select Customize display, then choose Search results. In the Select field to display as titles, choose document_id from the dropdown and click Apply. The document identifier in each search result will switch from the document name to the document id number.

When you are ready to start over, click Restore project defaults on the Improve and Customize page to restore the project defaults. {: tip}

Bonus: Handy tools at the top

{: #handy-tools}

  • To open the documentation, select the Help Help icon icon.
  • To view the available storage space and the number of documents and collections in your environment, select the Environment details Environment details icon icon. To delete all the collections in your environment, choose Delete environment.