Summary: Watson is a remarkably flexible and complete AI development platform. To understand how you might build new services for your current employer or imagine your own Watson-based startup, look at these 30 companies that are leading the way.
In our recent reviews of historical Watson and the modern Watson of today we concluded that IBM’s Watson Group may have the first or at least the current strongest comprehensive AI platform. This is the first time that we know of that all three elements of AI have been brought together in a single user friendly platform: image processing, text and speech processing, and knowledge retrieval.
This is not so much a platform for data scientist to use to expand the capabilities of AI as it is a platform for business users (with the aid of data scientists) to exploit the capabilities of modern AI by building new products and services.
To wrap up this review of Watson, we wanted to provide some thought-starters on what new services or even new businesses you might build on Watson.
Oh, and regarding new businesses, did we mention that developers who join the Watson Ecosystem are eligible to become a Watson “partner” with a shot at the $100 Million funding IBM is making available to startups plus support and access from IBM business and technology advisors.
Defining the Opportunity in General
Deciding whether or not Watson is right for your idea is simple. The problem you define will ideally require all four of these capabilities:
Another way to look at it is from the component standpoint.
The Knowledge Base
There must be some reasonably large body of knowledge you possess that is sufficiently complex that it’s difficult for any one person to fully grasp.
A person must need the information and be asking the question. If this was machine-to-machine there would be no need for the NLP processing and the questions would presumably be much less complex.
On the simple end, this might be all the possible rules and variations on how to set up a new account in a bank or how to order or return a product on your ecommerce site. On the complex end this may be all the possible symptoms, diagnoses, and procedures for a medical condition or potential failure and repair procedures for a complex device. On the really complex end, this could be all the known chemicals and how science allows them to interact in multiple combinations.
You will need to load and maintain the Knowledge Base, adding material as policies or possibilities expand and removing knowledge that is no longer accurate. In other words the Knowledge Base must be curated.
In a customer service application this may not be as difficult as you think since it could be loaded and largely trained directly from prior CSR logs and text-translated recordings. However, deciding what should be in the Knowledge Base and getting it there will be the most demanding part of the task.
The information contained in your Knowledge Base can have some very dynamic elements. For example, it might contain a constant social media stream (curated only to remove older material that is probably no longer relevant) allowing it to respond about trends perceived by some portion of the public.
Your Knowledge Base is not limited to text. It can also include audio clips, video, and still images. For example if the response to a query is best explained by a picture, or conversely if your user wants to input a picture and have you identify some appropriate action about that picture.
QAMs have the capability to fully interpret natural language whether written or spoken. More importantly they can also intake still images and videos as part of the query. For example, here is a picture of a rash on my arm – what should I do about it.
Outputs can also be combinations of text, speech, image, or video. Importantly, don’t forget augmented reality devices as an output. For example, in a preventive maintenance situation, the augmented reality device could display the exact repair steps needed displayed directly on the machine under repair, showing the technician exactly where to work and modifying its recommendation as each step is completed and fed back to the QAM.
Some Idea Starters
Sometimes it’s just easier to imagine how you can use a new technology if you have examples of how others have used it. So here are a number of short descriptions of new businesses or new services that have been built on Watson to get your creative juices flowing. You can see even more here.
Healthcare (Probably no area is more impacted by Watson than Healthcare)
1. Welltok built the CaféWell Health Optimization Platform which provides incentives for consumers to take care of their health. Welltok is IBM Watson’s first Ecosystem partner in consumer health and also IBM Watson’s first investment. User can ask questions in regular language, and get intelligent responses on health issues and fitness.
2. Medtronic PLC a manufacturer of medical devices will be using Watson's technology to predict diabetic attacks. The app helps patients monitor glucose levels by measuring the calories burned and food eaten.
3. Celgene Corporation and IBM Watson Health are co-developing IBM Watson for Patient Safety, a new offering that aims to enhance pharmacovigilance methods used to collect, assess, monitor, and report adverse drug reactions. Watson’s cognitive computing engine continuously learns, so it is expected that Watson for Patient Safety will increasingly be able to help identify potential drug safety signals.
4. Siemens Healthineers and IBM’s Watson Group are co-developing a platform for Population Health Management. The alliance aims to help hospitals, health systems, integrated delivery networks, and other providers deliver value-based care to patients with complex, chronic and costly conditions such as heart disease and cancer.
5. Talkspace is a global online platform that allows users to chat with a licensed therapist confidentially and anonymously. Talkspace is using IBM Watson’s Personality Insights API to better match users with therapists in their network using a self-learning system that seeks to better understand the traits of individual users.
6. LifeLearn used Watson to create Sofie, a tool to help Veterinarians diagnose and treat patients, a natural extension of Watson’s success in applications that diagnose and recommend treatments for human ailments. James Carroll, CEO of LifeLearn describes Sofie as "a treatment support application that provides veterinarians with the ability to derive evidence-based hypotheses with natural language inquiries to challenging medical situations."
Retail and E-commerce
7. Macys has developed “Macy’s On Call,” a mobile web app that taps Watson via the Satisfi API, an intelligent engagement platform, to deliver a first-of-its-kind solution that will enhance the customer in-store shopping experience. The app allows customers to input natural language questions regarding each store’s unique product assortment, services and facilities and receive a customized response to the inquiry. For example, a customer could type, “Where are the ladies shoes?” or type a combination of brand and product inquiry such as “INC dress,” and they will receive the relevant response and location of that product in the store.
8. The North Face, supplier of outdoor apparel and equipment uses Watson to enhance customer engagement. Customers can now use natural conversation as they shop online via an intuitive, dialog-based recommendation engine and receive outerwear recommendations that are tailored to their needs. Utilizing Watson’s natural language processing ability, XPS helps consumers discover and refine product selections based on their responses to a series of questions. For example, after a shopper enters details on a desired jacket or outdoor activity, XPS will ask questions about factors like location, temperature or gender to provide a recommendation that seeks to meet the shopper’s specific usage and climate needs.
9. VineSleuth developed its Wine4.me app to provide wine recommendations for consumers based on sensory science and predictive algorithms. The start-up uses Watson's language classifier and translation services in kiosks in grocery stores.
10. Decibel Music Systems is using IBM Watson to better collect and organize more qualitative data on musical influence and popular opinion. It collects these data points and trends to power its MusicGeek app, which helps users discover new music.
11. Reflexis Systems, Inc. is working with IBM Watson to enable retailers to uncover hidden demand triggers to deliver the best customer experience and prevent out of stocks. Reflexis StorePulse™ takes insights from IBM Watson and provides prioritized alerts and tasks with best practice actions to corporate, regional and store managers - based on local event, social media, and other trends and events that impact demand - to accurately plan each and every store accordingly every day.
12. Under Armour Inc. is using Watson to coach users of its apps by learning more about their exercise patterns.
13. Triax Technologies created the Triax Smart Impact Monitor (SIM™) which is a wearable sensor embedded in headbands or skullcaps that track the force and frequency of head impacts during play. By utilizing the Watson language service, this device can factor in more diverse data sources to analyze sentiment and infer cognitive and social characteristics to provide a more holistic view of athletic safety and performance.
14. Spare5 is creating a cognitive app called “Watson Golf Pro” that leverages Watson’s deep learning, natural language, and vision capabilities to act as a personal caddy that amateur players can consult while at the driving range or on the course. Based on knowledge about mechanics and drills obtained from contracted golf professionals combined with Watson’s ability to “see” a user’s golf swing will enable the app to provide feedback on how to implement better fundamentals. The Watson powered app can also deliver information and feedback in the right tone based on learning the nuanced personality of the users to better motivate the player to learn and take action.
15. 113 Industries is working with the Pittsburgh Penguins to transform the game-day experience for hockey fans. By leveraging 113 Industries’ “Pi” service embedded with Watson natural language cognitive capabilities, the Penguins can analyze large volumes of fan-based data to develop specialized offers and services for fans visiting the CONSOL Energy Center.
16. Edge Up Sports has become an IBM Watson Ecosystem Partner and is developing an app powered by Watson to help fantasy football team owners manage their team’s performance. "Edge Up grabs vast amounts of available NFL data, and with the help of Watson, users are able to make informed decisions and adjustments to their fantasy football roster picks,” said Edge Up Sports CEO Illya Tabakh. “Edge Up’s mobile app will provide core mental, physical and situational insights on NFL players, which pulls from social media interactions, weather reports, injury history, analyst write ups, news stories and more.
17. SoftBank is using Watson's technology in its Pepper robots to help it recognize customer emotions and use it as customer service staff.
18. Swiss Re is developing a range of underwriting solutions that rely on Watson. One of the first applications will be in Swiss Re's Life & Health Reinsurance business unit. Insurers must identify and act on emerging trends and operational issues or opportunities in real time and respond proactively. Cognitive technologies, coupled with human experience and insights, can enhance and help inform timely decision making. By applying Watson’s capabilities, the new platform could allow Swiss Re professionals to make better informed decisions and more accurately price risk.
Personal and Professional Services
19. UnitesUs, an online service for matching job seekers with employers, is using Watson's "personality insights" service. It amounts to automated, online personality assessments. The person looking for a job fills out a brief online application which asks for permission to mine their public Twitter messages, Facebook posts and other social-media writing. The Watson personality-assessment service then generates a report showing the hiring company information about the personality fit. About 7.5% of job seekers on UnitesUs have been called for interviews by employers compared to a typical 1% to 4% rate on unaided sites like Monster and Careerbuilder.
20. Case Strategy, a consulting firm, is using Watson to read, sort and assess vast amounts of financial, market and competitive data — from public government information, private sources and online social networks — to suggest growth opportunities for corporate clients.
21. ROSS Intelligence Inc. is using Watson to make it easier to get answers to legal questions. Users can ask questions in plain English and the app uses NLP to understand the questions and then sifts through the entirety of a database to return a cited answer with relevant legislation. ROSS also monitors potential changes to relevant laws and alerts you when changes occur.
22. InspireOne Technologies, a talent development service uses Watson to analyze employee e-mails to let them know what sort of leadership skills they display on a daily basis.
Building and Construction
23. Mears Group, the UK housing and social care provider has over 700,000 social homes to maintain across the UK and 20,000 employees. It also has over 600 million health and safety/accident reports on their database. Mears Group uses Watson to gain faster reporting turnaround and new insights and revenue opportunity suggestions.
24. ENGEO provides services such as environmental engineering and geotechnical engineering. To assist engineers in the field, ENGEO built its GoFetchCode app with Watson to help answer tough questions and bring information when it is needed. The app is especially useful in times such as after a natural disaster, when expertise around infrastructure is vital. GoFetchCode uses the cognitive capabilities of Watson and answers natural language questions for users.
25. ISS Company, a global provider of facility services, is using Watson IoT to transform the management of over 25,000 buildings around the world. Using data from millions of devices and sensors embedded into buildings including doors, windows, chairs, meeting rooms, dispensers and air conditioning systems. Data will be uploaded onto IBM’s Watson IoT cloud platform and cognitive computing technologies will learn from this data helping ISS optimize its services as well as furthering its understanding of how people use buildings. For example, sensors in doors and entrance areas can tell a real estate manager how many people are in a building at any one time and sensors on plate dispensers can inform kitchen staff of how many people are still likely to need to eat – helping staff to prepare the right amount of food and avoid waste.
Hospitality and Tourism
26. Hilton Worldwide uses Watson to power “Connie”–the first Watson-enabled robot concierge in the hospitality industry. Connie draws on domain knowledge from Watson and WayBlazer to inform guests on local tourist attractions, dining recommendations and hotel features and amenities.
Government and Policing
27. Imperial College London students created an app to predict crime with Watson.
28. Singapore's Inland Revenue Authority is using Watson to help answer questions about tax.
29. Purple Forge developed a Watson based 311 Service for Surrey, Canada to answer citizens' questions about government services. (When is recyclables pickup?) The app can answer more than 10,000 questions, more efficiently and at lower cost than humans.
30. Findability Sciences operates in the nonprofit arena. Their Watson based app “Impact Measurement & Analysis” gives donors an easy way to ask questions about potential investments. In particular, it gives donors a way to measure the impact of specific organizations which could entice them to donate multiple times.
Want to Cash In Personally on Watson?
There are many ways as a data scientist or developer to get up to speed on Watson however IBM has a formal agreement with Udacity to offer a new artificial intelligence (AI) Nanodegree program. According to the Udacity website this is a two-term, 26 week course covering logic and planning, probabilistic inference, game-playing / search, computer vision, cognitive systems, and natural language processing (NLP) among other topics.
About the author: Bill Vorhies is Editorial Director for Data Science Central and has practiced as a data scientist and commercial predictive modeler since 2001. He can be reached at: