Build an application powered by Watson (Part 5 of 5 Series)

Series: <<1 >><<2>><<3>><<4 >> <<5 >>

Follow IBM Watson -

Companies can embed Watson’s cognitive capabilities in their application without the need for building deep natural language, machine learning and ranking algorithms or other core technology skills. They could accomplish this by being able to embed the

Watson platform capabilities as a service, through the use of API and tools on the Watson Developer Cloud (WDC). The WDC is a part of the Watson Ecosystem, along with the Watson Content Store (WCS) and the Watson Talent Hub (WTH).

Could Watson be right for you? How to proceed?

The thought process behind embedding Watson starts with asking three simple questions –

1) Is there business value to this Watson powered application? To answer the question you need to evaluate the factors in creating a transformative application or significantly enhancing an existing application, pursuing a sizeable target market in a scalable manner, identifying key value propositions or differentiators, and being able to build a roadmap to revenue and profits.

2) Is there technical alignment between the application and Watson? Find out a good use case to leverage Watson's natural language and cognitive capabilities, using IBM's cloud.

3) Are there business and technical considerations on acquiring or maintaining content? Understand how to get access to usable unstructured content that can be acquired and maintained in a scalable manner. The availability of free or fee-based content in the WCS may also be considered.

In addition to the capabilities that Watson offers, there may be other products in the IBM portfolio that could lend complementary value to the solution, such as analytics, Big Data products or Business Rules systems.

Companies submit a request and follow a methodology to get access the WDC in the Watson Ecosystem. Initially, IBM expects that the ecosystem will start with three verticals – retail, healthcare and travel - but it will follow a “BYOV” (Build your Own Vertical) model – where, with the right content and applications, new verticals can be developed. Once access is obtained to the WDC, the capabilities of the Watson platform can be embedded in an app using tools and an Application Programming Interface (API) in one of two ways –

1. Using a customizable User Interface

A prepackaged IBM hosted user experience could be integrated within the customer application via an HTML inline Frame (iframe). An iframe can place another HTML document in a frame. This frame can be customized to look and feel like the target application.

2. Application Programming Interface (API)

The Watson Question and Answer API (QAAPI) is a Representational State Transfer (REST1) service interface that allows applications to interact with Watson. Using this

API, one can pose questions to Watson, retrieve responses, and submit feedback on those responses. In addition to simple question and responses, Watson can provide transparency into how it reached its conclusions through the REST services. The QAAPI supports JSON as the content-type and accept type.

There are other functions like ingesting content in the Watson platform that will also be exposed as tools and APIs and can be accessed from within an application.

Watson Question and Answer API (QAAPI)

QAAPI enables developers to post questions to Watson and retrieve responses and supporting evidence. The evidence contains supporting information based on the passages in the documents that Watson ingested and may have been trained with.

Watson supports two ways of using the QAAPI – Asynchronous and synchronous mode. With asynchronous mode, Watson also includes URLs for retrieving status and answers. In synchronous mode, there is no link URL.

1) Asynchronous mode:

In the asynchronous mode, the question is posted to Watson and the response returns a link to retrieve the answer when it is ready. The server needs to be polled to check if the question has completed processing. A status field in the response will indicate if the question contains the final answer. This mode can also be used if a bulk set of questions need to be submitted to Watson, with the answers being retrieved later. See the sample request & response headers here.

2) Synchronous mode:

In the synchronous mode, a POST operation sends the question to Watson. The answer is received synchronously. This eliminates the need for a polling logic in the client application. In order to make the request synchronous the following HTTP Header needs to be provided:

X-SyncTimeout : <time in seconds>

If this Header is absent, asynchronous mode is assumed. See the sample request & response headers here.

The <time in seconds> refers to the length of time the service should wait before giving up on a response. Thirty seconds is typically used as the default value. A value of -1 indicates the server should wait indefinitely. The timeout value does not refer to how long the browser or client application should wait, but rather how long processing time on the server should take to answer the question, before it times out.

Follow these examples to learn how to submit questions
and handle responses with the Watson Question and Answer REST service.

• Example simple post in synchronous mode
• Example post for answers with evidence
• Example post to control the number and format of answers
• Example for resubmitting a question with synonym feedback

The QAAPI uses Basic Authentication over SSL to provide security. During registration to the Watson Developer Cloud, a user ID and password are provided. This user ID and password is used for basic authentication.

Here’s the flow that an application component would use to access Watson Question and Answer capability using the QAAPI. The key steps involved are:

1. Configure parameters, including authentication

2. Post question

3. Receive response

4. Process response

Creating Cognitive applications Powered by Watson

What does it mean for an application partner to embed Watson’s capabilities? How is it done? Let’s take a case of a fictitious company “Think Travel” wants to build an application for consumers: the FictionTravel App. By embedding the capabilities of Watson in the application, FictionTravel can allow a user to imagine, plan, and book a vacation through an interactive conversation, all while staying in the same application.

Think Travel can extend its current capabilities beyond booking flights and hotels by utilizing Watson's cognitive capabilities to improve user experience. Watson could transform how users could book travel by offering an app that gives tailored advice through conversation, similar to a travel agent. It could provide a resource for a user to go through an exploratory phase of their travel options from a centralized location using trusted information. One customer may ask, “I want to take a week-long trip with my family to the beach in early September. Where is the best place to go?” Consequently, Watson will answer and continue the dialogue, providing possible considerations and engaging the user to get to an answer.

While assisting the user with their travel needs, it will learn about the user and provide increasingly contextual responses, helping the user narrow down the location that might be a good fit. Watson can address a myriad of concerns from booking a hotel to visa requirements, all without ever leaving the application. This will create a simple and enjoyable planning process for the user.

Not only is it simple for the user, but also for the developers. The cognitive application is built using capabilities of the Watson platform without the app developers having to acquire a new complex set of technology skills on cognitive systems. App developers are able to completely transform the user experience using Watson's cognitive capabilities and intuitive tools in their application by accessing an API. Separately, the business and domain experts at Think Travel will identify and obtain the content for Watson.

On the back end, Watson is doing the complex processing to sift through thousands of travel documents, country immigration policies, destination information, books, travel guides, user reviews, and more to infer information, interpret user questions and provide meaningful responses throughout the conversation. Learning from customer interactions, it can give the user a seamless travel experience through the FictionTravel app.

 

Overall, Watson is able to do the heavy lifting to provide easy access and high value for developers, customers and the business as a whole. Imagine having an expert travel adviser in your pocket. That is how FictionTravel App, powered by Watson, can positively disrupt customer behavior. Please refer this whitepaper for more details “An Ecosystem Of Innovation – Creating Cognitive Applications Powered By Watson”.

Reference:

• Using the Question and Answer API to Access Watson Capabilities
• Overview of the Question and Answer REST Service
• Application development flow
• Examples with the Question and Answer REST service
• Operations on Question: POST /deepqa/v1/question
• Partner applications powered by Watson

Whitepapers –

• An Ecosystem Of Innovation – Creating Cognitive Applications Powered By Watson
• Creating Cognitive Apps Powered By IBM Watson – Getting Started with the API
• The Era of Cognitive Systems – An Inside Look at IBM Watson and How it Works
• Transforming the Way Organizations Think with Cognitive Systems

Videos -

• IBM Watson Group Launch
• What will you do with Watson?
• Watson fueling new era of apps
• Science behind Watson
• Watson oncology demo
• What is Watson?
• A look inside IBM Watson solutions lab

Watson for ISVs/Partners and entire Ecosystem (Part 4 of 5 Series)

Series: <<1 >><<2>><<3>><<4 >> <<5 >>

Follow IBM Watson -

So far in this series we saw how IBM’s Watson System is at the forefront of cognitive computing. By saying that I mean that Watson processes a question through a similar approach that humans do. It starts by analyzing the question (or case) as input and generates a set of features. It then generates a set of hypotheses, by looking across passages from the consumed content, seeking the best potential response to the question. Now for each of the candidate answer it scores based on the relevance of the answer using hundreds of reasoning algorithms embedded in the system until the most probable answer and its corresponding evidence surfaces to the top, with the associated highest confidence.

In mid-November 2013 IBM announced the IBM Watson Ecosystem Program. As per the program, Watson technology is available as a development platform in the cloud, to enable a worldwide community of software application providers to build a new generation of apps infused with Watson's cognitive computing intelligence. Aim is to spur innovation and fuel a new ecosystem of entrepreneurial software app providers who will bring forward a new generation of applications infused with Watson's cognitive computing intelligence. The ecosystem is expected to attract organizations of all sizes, forms, geographies, industries, and specialties. It has the potential to fuel new entrepreneurial ideas and start-ups.

 The ecosystem approach poses many benefits for partners. ISVs can leverage IBM’s investments in developing Watson’s transformative capabilities to make their apps more valuable. Building on the momentum of past successes, IBM is assembling leading content providers, independent software vendors (ISVs), and talent providers to collaborate toward the development and release of “Powered by IBM Watson” applications. The initiative will leverage IBM’s extensive experience working with tens of thousands of business partners to help drive innovation. Good partner candidate is expected to have an established business model, a defined use case for Watson, internal resources available to execute the program and go to market strategy, and executive commitment. A good use case will leverage many or all of these Watson Capabilities:

• Exploits the 4 Vs of data–volume, velocity, variety and veracity
• Leverages evidence-based insights with weighted confidence
• Benefits from a continuously learning system
• Takes advantage of deep natural language processing
• Requires transparency to the source of information
• Transforms user experience with contextual relevance and active dialoguing

IBM offers a single source for developers to conceive and produce their Powered by Watson applications.

Watson Developer Cloud (WDC) will offer the technology, tools and APIs to ISVs for self-service training, development, and testing of their cognitive application. The Developer Cloud is expected to help jump-start and accelerate creation of Powered by IBM Watson applications.

Watson Content Store (WCS)

The fuel that keeps the Ecosystem running is information. Quintillions of bytes of information are created every day, providing the basis for new insights and possibilities. WCS is intended to bring together unique and varying sources of data, including general knowledge, industry specific content, and subject matter expertise, to inform, educate, and help create an actionable experience for the user. Each application is expected to be defined on the WDC by partners brings their own content and draw from the WCS.

WCS brings together unique and varying sources of data, including general knowledge, industry specific content, and subject matter expertise to inform, educate, and help create an actionable experience for the user. Content can come from a variety of sources, including the ISVs themselves, IBM or a third party. The store is intended to be a clearinghouse of information presenting a unique opportunity for content providers to engage a new channel and bring their data to life in a whole new way.

Watson Talent Hub

Talent providers and hubs will help ensure the availability of the human skills that it will take to achieve success of an app in the market. These individuals will work and assist in the creation of ground-breaking inventions.

Staffing and talent organizations with access to in-demand skills like linguistics, natural language processing, machine learning, user experience design, and analytics will help bridge any skill gaps to facilitate the delivery of cognitive applications. These talent hubs and their respective agents are expected to work directly with members of the Ecosystem on a fee or project basis. 

Partner roles in the Ecosystem:

Application Partners

For example, through the Watson Developer Cloud, a consumer-oriented application could deliver a personalized guided selling experience online that can replicate the best in-store experience. Or, an airline could create virtual travel agents able to recommend the perfect destination tailored to the user’s unique needs.

Content Partners

Organizations that have the rights to use and/or license general and domain specific content are ideally suited for the program. Information based organizations like publishers, researchers, and even social forums could take advantage of the Watson Content Store. Conventional data sources, representing millions of pages, including reports, publications, user-group logs, clinical trial data, call center recordings or product manuals could help fuel a new generation of cognitive apps.

Talent Partners

IBM anticipates offering a talent certification program through a range of Talent Partners participating in the ecosystem. IBM anticipates offering one or more levels of certification. 

The partner responsibilities within the Watson Ecosystem are summarized below:

5 Steps for ISV/Partner to a Powered by Watson application and profitability

1. Identify and develop your use case where embedding a Watson capability into an application presents a unique value proposition.
2. Identify content or content licensed through the WCS to support your application.
3. Design and develop your offering using the WDC to gain access to your sandbox instance of Watson.  The Watson Experience Manager (WEM) is a cloud based tooling interface to upload and ingest content, manage your environment, and work with Watson.
4. Train and test the offering through the WEM’s training and user experience configuration tools.
5. Deploy and roll out through IBM’s hosting solutions while leveraging application metrics to refine and evolve your solution.

I am interested. How do I apply for Watson? Follow the link and request the access:

https://www.ibmdw.net/watson/request-access/

Here are few Partner case-studies -

1.      Fluid Inc. Case Study 

2.      MD Buyline, Inc. Case Study

3.      Welltock, Inc. Case Study

Reference:

1. Brochure can be downloaded from here.
2. Watson Whitepapers:

• An Ecosystem Of Innovation – Creating Cognitive Applications Powered By Watson
• Creating Cognitive Apps Powered By IBM Watson – Getting Started with the API
• The Era of Cognitive Systems – An Inside Look at IBM Watson and How it Works
• Transforming the Way Organizations Think with Cognitive Systems

Watson’s Fitment For Today’s Challenges & Usecases (Part 3 of 5 Series)

Series: <<1 >><<2>><<3>><<4 >> <<5 >>

Follow IBM Watson -

A combination of capabilities makes Watson unique, including natural language processing, hypothesis generation and evaluation, and dynamic learning. Watson is about bringing these capabilities together in a way that’s never been done before resulting in a fundamental change in the way businesses look at quickly solving problems

• Solutions that learn with each iteration
• Capable of navigating human communication
• Dynamically evaluating hypothesis to questions asked
• Responses optimized based on relevant data
• Ingesting and analyzing Big Data
• Discovering new patterns and insights in seconds

Before we talk about what Watson can do, let us understand today’s scenario in a better manner thru below facts and figures -

• “90% of the world’s data has been created in the last 2 years.”
  Source: GigaOM, By Nicole Solis Mar. 23, 2011

• “80% of the world's data is unstructured.” 
  Source: Multiple reports reinforce this point IDC, Gartner Jan 2010, IDC 2012, McKinsey

• “In 15 of the US economies 17 sectors, companies with more than 1,000 employees store, on average over 235 terabytes of data -- more data than is contained in the US Library of Congress.”

Source: McKinsey & Company - Are you ready for the era of big data?  October 2011

• Natural Language Question Answering is a transformational benefit that will be adopted by the mainstream in the next 5-10 years.
  Source: Gartner July 2011 priority matrix for emerging technologies.

• “Total amount of global data is expected to reach 2.7 zetabytes in 2012, up 48% from 2011.”
  Source: IDC, December 2011 press release
  This is equivalent to….
• ~750 Trillion Hours of Video
• 50M times the amount of information contained in all the books ever written
• Every person on earth receiving 240 newspapers per day
• Loading up CD’s that would stack from the moon and back 5.5 times
• Total amount of global data is expected to reach 35 zetabytes by 2020, growing 44X over the decade.

• Social media and the web has been a catalyst in the information explosion
• 220 Billion pieces of user generated content exists on the web today
• 200M tweets are sent daily
• 249B e-mails (2.8M/sec)
• 35 hours video uploaded to YouTube per minute

• “1 in 2 Business leaders don’t have access to data they need.”
  Source: MIT Sloan Management / IBM IBV Study, New Path to Value

• “83% of CIO’s cited BI and analytics as part of their visionary plan.”
  Sources: The Essential CIO 2011 Report / IBM Global CIO Study

• “5.4X more likely that top performers use business analytics.”
  Source: MIT Sloan Management / IBM IBV Study, New Path to Value

• “Analytically sophisticated organizations are 260% more likely to be top performers than analytic beginners.”
  Source: MIT IBV study

Above points indicate that the data is growing at an astounding rate.  It is growing so fast that we often lack the ability to use it to its full potential.  The highly unstructured nature of this data makes the challenge even more difficult.  This is a real problem for business to make informed decisions.  Business leaders need a way to find hidden patterns and isolate the valuable nuggets that they need to make business decisions.

Challenges are real and Watson has the answers. Watson use cases can be broadly be broken into three classes – Ask, Discover, and Decide. Lets see the details about each class and figure out under which class your application/solutions falls in. 

• Users can ASK Watson direct questions in natural language the same way they ask friends or colleague’s questions.  This is in contrast to reducing an inquiry to a set of keywords and receiving a set of links to sources where their answers may or may not lie.  People who saw Watson’s victory on the quiz show Jeopardy! will be familiar with this simplest use case. Think of this as next generation chat. 
• Users can DISCOVER new insights with Watson.  Examples of this could be use of Watson as a research assistant such as a biotech investigator looking for the best way to treat a disease in a specific cohort of patients. 
• Finally, users might use Watson to help them DECIDE on the best course of action.  This would be for situations where users are looking for confidence-based recommendations for their next action when they have many options to choose from such as what course of treatment to prescribe to a patient or what investment choice to make. 

Watson applies human-like characteristics to conveying and manipulating ideas. Please check out the figure below to understand key elements of Watson and download The Era of Cognitive Systems: An Inside Look at IBM Watson and How it Works for detailed description.

Watson is well-suited for a wide variety of applications. Potential applications can be loosely grouped into the following categories:

• Diagnosis and action - Assistance for knowledge workers dealing with a single case for a single client to pinpoint a condition from among many possibilities and make resultant decisions
• Contact center support - Personalized self-service experience for clients by dynamically developing personal profiles from unstructured data
• Research and discovery - Identification of rare studies and information sources while building a case for original research
• Process optimization - Identification of areas for improvement in business processes by analyzing unstructured data that documents and describes process steps and output
• Fraud and risk management - Identification of early signs of fraud and management of risk in order to lower overall liability and costs of doing business

Prev: The science and technology behind IBM Watson (Part 2 of 5 Series)
Next: Watson for ISVs/Partners and entire Ecosystem (Part 4 of 5 Series)

References -

Watson is ready for Business

http://www.ibmwatson.com/
IBM Watson
www.ibm.com/innovation/us/watson/index.shtml
Watson YouTube channel
http://www.youtube.com/user/IBMWatsonSolutions/feed
Watson in Google News
https://news.google.com/news/search?hl=en&gl=us&q=ibm+watson
IBM Journal - This is Watson
http://ieeexplore.ieee.org/xpl/tocresult.jsp?reload=true&isnumber=6177717
Ventana Research: The Potential of Cognitive Computing Platforms
http://www.ibm.com/innovation/us/watson/pdf/ventana_research_perspective_cognitive_platforms_ibm.pdf
Join Watson Community @ IBM developerWorks

https://www.ibm.com/developerworks/community/groups/community/Watson

The Science And Technology Behind IBM Watson (Part 2 of 5 Series)

Series: <<1 >><<2>><<3>><<4 >> <<5 >>

Follow IBM Watson -

DeepQA is the underlying architecture that’s powering Watson. Actually DeepQA is software architecture for deep content analysis and evidence-based reasoning that embodies that philosophy. It represents a powerful capability that uses advanced natural language processing, semantic analysis, information retrieval, automated reasoning and machine learning. DeepQA deeply analyzes natural language input to better find, synthesize, deliver and organize relevant answers and their justifications from the wealth of knowledge available in a combination of existing natural language text and databases.

The DeepQA architecture views the problem of Automatic Question Answering as a massively parallel hypothesis generation and evaluation task. As a result DeepQA is not just about question-in/answer-out – rather it can be viewed as a system that performs differential diagnosis: it generates a wide range of possibilities and for each develops a level of confidence by gathering, analyzing and assessing evidence based on available data.

With a question, a topic, a case or a set of related questions, DeepQA finds the important concepts and relations in the input language, builds a representation of the user’s information need and then through search generates many possible responses. For each possible response it spawns independent and competing threads that gather, evaluate and combine different types of evidence from structured and unstructured sources. It can deliver a ranked list of responses each associated with an Evidence Profile describing the supporting evidence and how it was weighted by DeepQA’s internal algorithms.

“Natural Language Question Answering is a transformational benefit that will be adopted by the mainstream in the next 5-10 years.”
Source: Gartner July 2011 priority matrix for emerging technologies.

In short, powered by DeepQA, Watson is the combination of these capabilities -

Natural language processing by helping to understand the complexities of unstructured data, which makes up as much as 90 percent of the data in the world today

Hypothesis generation and evaluation by applying advanced analytics to weigh and evaluate a panel of responses based on only relevant evidence

Dynamic learning by helping to improve learning based on outcomes to get smarter with each iteration and interaction

Let us see what technological pieces make Watson a reality. The Jeopardy! Watson system configuration included 2880 processing cores, 16TB memory, 20TB disc and could process at 80 teraflops per second. Watson demonstrated its ability to sift through an equivalent of about 1 million books or roughly 200 million pages of data, and analyze this information and provide precise responses in less than three seconds. For Jeopardy, Watson held nearly twice the information contained in the US Library of Congress in instantly accessible RAM memory. On the software side, its information management capabilities combine to deliver deep content analysis and evidence based reasoning that connect widely disparate sources of information and make the kinds of connections that humans make.

Watson is 240 times stronger now in terms of processing capabilities after Jeopardy!. Capabilities like content analytics, business analytics, big data, database and data warehouses are areas where IBM has led the market for years and which form the backbone of Watson’s strength. 

For more details please download below eBook –

The Era of Cognitive Systems: An Inside Look at IBM Watson and How it Works
http://www.redbooks.ibm.com/abstracts/redp4955.html

Prev: What is IBM Watson? (Part 1 of 5 Series)
Next: Watson’s fitment for today’s challenges & use-cases (Part 3 of 5 Series)

 References -

Watson is ready for Business
http://www.ibmwatson.com/
IBM Watson
www.ibm.com/innovation/us/watson/index.shtml
Watson YouTube channel
http://www.youtube.com/user/IBMWatsonSolutions/feed
Watson in Google News
https://news.google.com/news/search?hl=en&gl=us&q=ibm+watson
IBM Journal - This is Watson
http://ieeexplore.ieee.org/xpl/tocresult.jsp?reload=true&isnumber=6177717