Many of you have used the IBM Watson Question and Answer (QA) service that allows you to ask questions and get answers on Bluemix, and have asked how and when it will be possible to bring in your own data.

This summer we released four new services that will replace the Watson QA service and help you customize and embed question and answer capabilities into your application. The new services can be trained on your industry or application specific data, and do not require deep knowledge in machine learning and linguistic models. As a result, the Watson QA service tile will be removed from the Bluemix catalog on Friday, November 20, 2015, after which you cannot provision new instances of this service. However, existing provisioned instances will continue to be available until December 16, 2015.

The four replacement services recommended for different types of question and answer capabilities are:

classifier-64Natural Language Classifier – allows you to Interpret and classify natural language with confidence.

Dialog_Icon-60pxDialog – allows you to script a conversation and help walk a user through a process

retriveandrankRetrieve and Rank – allows information retrieval with a machine learning model

passage-prep-160Document Conversion – takes documents and ‘chunks them up” into smaller answer units to return as passages


When trying to build a question and answer system there are two different starting points which ultimately can help you build a comprehensive and robust system.

One approach is to start building out a system that is able to answer the most highly repeatable questions with defined answers. For this use case we recommend starting with a combination of Watson Dialog and Natural Language Classifier. You can merge these services together with speech services to create a voice based question and answer system. This will allow you to answer a large percentage of your user’s questions in natural language, and also enable you to learn from usage what other questions you need to train the system on. We have a few demo applications to get you started in the App Gallery:  What’s in Theaters and Questions on the Natural Language Classifier are great starting points.
NLC and Dialog

Another approach is to build a question and answer system based on documents such as FAQs or marketing materials. In this scenario you would use Document Conversion to transform your documents into Answer units, and use those as input to train the Retrieve and Rank service. As in the first use case, you will learn from usage how and what your users are asking, and use that as input to improve the system.
RR and DC

Either of these approaches can help you get started in creating a full fledged question and answer application.

Also check out these recent blogs on applying the new “QA building blocks” services:

We would love to hear from you! If you have any comments or questions, please reach out to us in the Watson Developer Forum, or leave a comment below.

7 comments on"Watson Question and Answer service to be withdrawn"

  1. Michel Jonker November 12, 2015

    Are there any limitations with regard to the Natural Language ?

    In other words, can I classify the language to be Dutch and also covert Documents that are written in Dutch ?

    If I wanted to recreate this, can I do this with IBM services only ?

    • Rahul A. Garg November 12, 2015

      As to limitations would need to understand more on that question. There is work going on to add more languages. Currently NLC only supports English stay tuned in coming month we will be introducing more languages. You could use Watson Langage Translation Service to try to recreate solution referenced in the Developerworks article.

  2. Watson Q@A was one of the very first Watson services, and was available on Bluemix all the way back in the beta period. I was shocked to discover recently that it was still *in* beta, considering that almost all of the other services, despite having been created more recently, have far surpassed it. Meanwhile, the Watson Engagement Adviser, offering a very similar services, has advanced continuously.

    Reading this now, I see that IBM itself was unsure of what to do with the application. It’s good to have a resolution, finally, to this conundrum. I do not think that these new services, replacing one platform with four, are going to be as robust as Q&A was meant to be, but I can see Q&A being to monolithic an idea to work well.

    • Rahul A. Garg November 17, 2015

      We believe the services outlined will make it easier for developer to integrate QA into their applications and customize QA fairly quickly. As to Robustness our team believes that the 4 services we identified are more advanced and extremely robust to provide a great solution for developers.

  3. Fernanda Perego November 21, 2015

    I heard in the past that some people within the CoC (Development) in Ireland were working in a project for multi-language support. Do we have to expect some results from this work or the project has been left and we need to count only on the Watson native language support to fully exploit a localized solution?

    • Rahul A. Garg November 21, 2015

      Work from the CoC has been incorporated into our Langage Support development for the Watson services. On 11/16/15 Natural Language Classifier added support for Spanish and Portuguese more information here you will see other services adding language support as well.

  4. Joey Richardson April 06, 2016

    My team and I are trying to move over from Q&A to the breakdown services of NLC, R&R, Document Converter, and Dialog. We were using a Watson ecosystem for the Q&A service and made rich formatted word documents for the Watson Corpus. In the Dialog tool documentation, it states it can use JSON unit answers as a dialog, but using the Document Converter to change our word document to JSON unit answers the dialog tool does not accept our JSON. This is the error:

    { code: 400,
    error: ‘Failed to import file. Possibly due to corrupt or invalid file or system error. – com.fasterxml.jackson.databind.exc.UnrecognizedPropertyException: Unrecognized field “source_document_id” (class, not marked as ignorable (13 known properties: “entities”, “constFolders”, “hitNodeRules”, “translations”, “settings”, “stopwords”, “profileValueRules”, “nodes”, “dictionaries”, “version”, “specialSettings”, “varFolders”, “reports”])\n at [Source:; line: 2, column: 26] (through reference chain:[“source_document_id”])’ }

    Can you provide a direction for us to look into?

Join The Discussion

Your email address will not be published. Required fields are marked *