Watson Services

What is Watson?

Watson is a cognitive technology that can think like a human. [source]

Watson will help you build cognition into your apps and products. To see what you can do, check out the summary below.

Step 1: Create a Bluemix account

IBM Bluemix is a cloud platform as a service (PaaS) developed by IBM.

Watson services will be run on top of Bluemix. Which means you will need a Bluemix account to get started. To get that setup, create your account here.

Step 2: Follow Tutorials

Once Bluemix has been setup, take a look at the tutorials for Watson services. They walk you through getting your credentials setup on Bluemix to configuring and making requests to your Watson service. For more details on the actual API calls, check out the API Explorer. If you have any questions, just ask any one of us.

Step 3: Connecting Watson Services to your App

There are two options to connect Watson services to your application.

Option #1:

You can make the REST calls directly by checking out the API Reference for each service or through the API Explorer.

Option #2:

Another option is to use the already provided libraries that allow you to interact with the Watson services. To get started with this, check out the libraries for your platform on Watson Developer Cloud's Github page.

Step 4: Additional Resources


Visit the site directly to get an overview of the Watson APIs or check out our summary below:

  • Alchemy API

    • Description: Uses NLP and ML algorithms to extract meta-data from content, such as information on people, places, companies, topics, facts, relationships, authors, and languages.
    • Input: Web pages, HTML content, or text content
    • Output: Metadata based on the content that was passed
    • Notes:
  • AlchemyData News

    • Description: Provides news and blog content enriched with natural language processing to allow for highly targeted search and trend analysis.
    • Input: A query with NLP to search both the text in indexed content and the concepts that are associated with it.
    • Output: News and blog content enriched with our full suite of NLP services. Keywords, Entities, Concepts, Relations, Sentiment, Taxonomy.
    • Notes:
  • Conversation

    • Description: Create a bot/virtual agent to allow end users to communicate naturally with your system
    • Input: Represent a template of your converstaiton in the form of intents, entities, and crafted conversations
    • Output: A trained model that enables natural conversations with end users
    • Notes:
  • Dialog

    • Description: Enables applications to use natural language to respond to user's questions or comments.
    • Input: Script conversations based on your expert knowledge of the domain.
    • Output: End users can chat with your application using natural language and get the pre-written responses you created.
    • Notes:
  • Document Conversion

    • Description: This service prepares documents so they can be used by other Watson APIs.
    • Input: A Microsoft Word Document, A HTML Document, A PDF Document
    • Output: An answer unit JSON document, a plain text document, or a HTML document
    • Notes:
  • Language Translation

    • Description: Translate text into one of the supported languages.
    • Input: Plain text in one of the supported input languages and domains.
    • Output: Plain text in the target language selected.
    • Notes:
  • Natural Language Classifier

    • Description: Handle common questions from users, classify SMS texts as personal, work, or promotional, classify tweets, and control outcome of user.
    • Input: Text to a pre-trained model
    • Output: Classes ordered by confidence
    • Notes:
  • Personality Insights

    • Description: The service outputs personality characteristics that are divided into three dimensions: the Big 5, Values, and Needs.
    • Input: JSON, or Text or HTML (such as social media, emails, blogs, or other communication) written by one individual
    • Output: A tree of cognitive and social characteristics in JSON or CSV format
    • Notes:
  • Relationship Extraction

    • Description: Analyze news articles and perform linguistic analysis of the input text.
    • Input: Text news articles
    • Output: XML document of the entities from the text and the relationships of said entities.
    • Notes:
  • Retrieve and Rank

    • Description: Find the most relevant information for your query using machine learning algorithms.
    • Input: Documents, Queries (Questions), and User Queries (Questions)
    • Output: Indexed Documents, Rank (ML Model), List of relevant documents and metadata.
    • Notes:
  • Speech to Text

    • Description: Converts speech into text.
    • Input: Streamed or recorded audio
    • Output: Text transcriptions of the audio
    • Notes: The transcription of incoming audio is corrected as more speech is heard. Supported languages include US English, UK English, Japanese, Spanish, Brazilian Portuguese, Modern Standard Arabic, and Mandarin.
  • Text to Speech

    • Description: REST API to convert text to speech in many different voices.
    • Input: Plain text in one of the supported languages.
    • Output: Returns the audio in ogg, wav, or flac formats.
    • Notes: Developers can control the pronunciation of specific words. Supported languages include Brazilian Portuguese, English, French, German, Italian, Japanese, and Spanish.
  • Tone Analyzer

    • Description: Detects three types of tones from text: emotion, social tendencies, and language style
    • Input: Text
    • Output: JSON object that represents the analysis of the input message.
    • Notes:
  • Tradeoff Analytics

    • Description: Helps people make decisions when balancing multiple objectives by using the Pareto Optimization filtering technique.
    • Input: A decision problem with objectives.
    • Output: JSON object that represents optimal options and highlights the tradeoffs between them.
    • Notes:
  • Visual Recognition

    • Description: Understand the content of images.
    • Input: Image
    • Output: Relevant classifiers related to objects, events, and settings.
    • Notes: Can also train the API to detect custom content.

results matching ""

    No results matching ""