Tag Archives: Natural Language Understanding

IBM Watson

An Immersive Experience With IBM Watson

IBM Watson is a cognitive intelligent system that uses natural language processing and machine learning to reveal insights from large amount of unstructured data. You may remember it for winning the quiz show Jeopardy in 2011.

On 22 June 2017, IBM invited us to participate in an exclusive demonstration of Watson’s capabilities. They also showed us how Watson allows even non-techies mine large amounts of abstract data for pertinent information.

For a similar demonstration of artificial intelligence and big data analytics capability, check out our recent article – The AWS Masterclass on Artificial Intelligence by Olivier Klein.

 

An Immersive Experience With IBM Watson

In development for over 12 years, IBM Watson is understandably far complex than when it debut in Jeopardy. It is powered by 50 underlying cognitive technologies, including natural language processing, machine learning and deep learning.

This particular session was conducted by Zainal Azman Shaari and Cheok Swin Voon from IBM’s Eco-system Development team, and is limited to the functions of IBM Watson Conversation and its natural language understanding (NLU) capabilities.

A Monash University team of final year students then showed what they accomplished with IBM Watson in just a few weeks, despite not having any programming abilities. Specifically, they showed how they can mine Twitter to not just determine how often people were talking about a fast food brand, they also used IBM Watson to determine whether the tweets were positive or negative in nature.

For a similar demonstration of artificial intelligence and big data analytics capability, check out our recent article – The AWS Masterclass on Artificial Intelligence by Olivier Klein.

Next Page > The IBM Watson Q&A Part 1/2

[adrotate group=”1″]

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!

The IBM Watson Q&A Part 1/2

  1. What is IBM Watson?
  • IBM Watson is a cognitive intelligent system that uses natural language processing and machine learning to reveal insights from large amount of unstructured data. It has the capability to tackle tough problems in every industry from healthcare to finance.
  • Through understanding the data it learns from human input, it is not only able to improve its response accuracy but also understand trends that can then be utilised to generate business insights.
  • Watson is powered by 50 underlying cognitive technologies – including natural language processing, machine learning and deep learningto provide capabilities that span language, speech, vision and data insights.

 

  1. Can you tell us about Watson’s history? Who are the brains behind its inception?
  • IBM has been researching, developing and investing in AI technology for more than 50 years. The rise of big data (2.5 billion gigabytes of data are created every single day) and the world’s need to make sense of it all – is what ultimately led IBM researchers to develop Watson.
  • When IBM Watson won the historic Jeopardy! exhibition on TV in 2011, it was a watershed moment. And since then we’ve only accelerated our innovation, advancing and scaling the Watson platform and applying it to many industries, including healthcare, finance, commerce, education, security and IoT. We’ve engaged thousands of scientists and engineers from IBM Research and Development, and partnered with our clients, academics, external experts and even our competitors, to explore all topics around AI.

  • IBM Research is today the largest industrial research organization in the world, with twelve labs on six continents, and the company invests $6 billion on R&D.
  • The labs are viewed as one of IBM’s greatest assets and a vital part of defining the future. In August 2016, we celebrated the 30th anniversary of the Almaden research lab. Today, Almaden’s research community focuses on solving problems across areas as diverse as nanomedicine, services science, atomic scale storage, food safety and medical image analytics.

 

  1. What are some of IBM Watson’s accomplishments in the real world today?
  • Since its debut, Watson solutions are now being built, used and deployed around the world and across industries, including healthcare, financial services, travel and retail. Since 2015, IBM has seen a 400 percent increase in Watson app creation on Bluemix, an increase of over tenfold over the last year.
  • Watson has played a role in supporting more than 200 million patients through healthcare systems in the U.S., China, Japan, Thailand and India, collaborating with doctors, helping improve treatment recommendations and helping deliver more efficient care.
  • Watson is also available to more than 200 million consumers to answer their questions, find what they need and make recommendations with greater personalization.
  • At the same time, Watson is tackling dozens of meaningful business and societal challenges, helping reduce building emissions, reducing hiring cycles and finding ways for cities to save more water and other resources. In education, half a million students can choose their courses to master a subject with Watson and we’re also helping teachers address each student’s unique needs.

 

  1. With the interaction between IBM Watson and regular users becoming increasingly more complex and utilize more of the user’s personal information, is there a potential risk for loss of privacy for users in an unfortunate incident?
  • While Watson is indeed able to process more forms of information, we have also ensured that we have developed safety measures within Watson as well as continue to work with our industry partners all to ensure that the data that Watson has access to is not accessible by any unauthorized third-parties.

 

  1. It was mentioned that IBM Watson understands natural languages. What does this mean and what languages does it support?
  • IBM Watson is able to understand sentences spoken by people by analysing the words used as well as the context of the sentence in order to accurately interpret sentences.
  • Currently IBM Watson is able to understand and reply in nine different languages including German, Spanish, Korean, Japanese and English to name a few. We are also constantly updating Watson with additional languages to allow more people in the world to experience him.
[adrotate group=”2″]
  1. Why is understanding natural human speech such a complex task for artificial intelligence systems to perform?
  • The primary difficulty when it comes to recognizing human speech is that natural speech is inherently filled with unstructured data, and this is the same across all forms of content written by humans for other humans.
  • The main difference is that normal structured data is governed by well-defined rules whereas speech is governed by rules of grammar, context and culture which is implicit, ambiguous and complex.
  • With a host of factors to consider and eliminate, recognizing speech can be both a time-consuming task as well as one that is open to a host of errors if not interpreted correctly.

 

  1. What is the systems implemented into Watson that allows him to easily comprehend human speech?
  • To facilitate IBM Watson to better tackle the challenge of understanding human speech, Watson is fitted with two different set of APIs that give it the ability to both comprehend and respond to human interactions: Natural Language Understanding (NLU) & Watson Conversation.
  • The Natural Language Understanding API allows Watson to analyze text to extract meta-data from content such as keywords, entities, sentiment, emotion, relations and many more.
  • Watson Conversation, on the other hand, functions as a visual dialog builder that helps create natural conversations for Watson that can be deployed on chatbots and other virtual agents.

Next Page > The IBM Watson Q&A Part 2/2

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!

The IBM Watson Q&A Part 2/2

  1. It was mentioned that Watson is capable of self-learning and improvements, can you explain more how this system works?
  • What makes Watson truly unique from pre-programmed AI is that Watson is capable of consistently learning from user input as well as correct itself to provide better answers to human queries.
  • Watson does this by undergoing a constant feedback loop with human input that involves correction of errors in order to determine the most appropriate answer that is based on the context of the question.

 

  1. What is the vision that IBM has for Watson and what is IBM currently doing to ensure that they can achieve this vision?
  • Our vision for IBM Watson is to develop it into a full-fledged platform that is capable of making a real impact in any industry it is applied in.
  • We are also focused on developing Watson as an Artificial Specific Intelligence, also known as narrow AI, that is specifically focused on relevant knowledge in its required field of expertise to ensure that they can respond in the same way professionals do.
  • We are currently actively working with professionals in the field as well as expanding the capabilities of Watson in order to ensure Watson is able to reach out to over 1 billion consumers by the end of 2017.

 

  1. Can you explain more about the term ‘machine learning’?
  • Machine learning can be defined as AI system that possesses the ability to learn without requiring constant programming.
  • Programs with machine learning systems are able to make use of most forms of new data it is exposed to develop and improve responses or solutions.
  • In future, AI systems will be able to understand more complex data while also delivering more accurate results compared to humans. This could allow future machines to be more autonomous while also handling challenging tasks that normal humans cannot do.

  1. In order to implement Watson, are developers required to pay a certain fee in order to utilize related assets and data?
  • No, interested developers who want to make use of the Watson platform to create their own cognitive apps or businesses can do so from the public Watson Developer Cloud and other platforms.
  • Our cloud platform represents the easiest way for potential developers to create their next-gen cognitive apps with a total of over 160 unique APIs and services in 179 countries.

 

  1. Does IBM foresee a drop in openings for jobs that Watson is able to perform given his increasing capabilities for complex tasks?
  • With every major advancement in technology, there will always be a change to the existing workforce where there will be new ways of working as well as new skills and jobs that will be in greater demand. This is the same case as well for cognitive computing.
  • Despite this, we believe that Watson and the field of cognitive computing will instead introduce a whole new level of collaboration between man and machine, and serve to expand human intelligence instead of replacing it.
  • We believe that by thoughtfully and purposefully combining the best qualities of both man and machine, we can understand our world better as well make better decisions that can lead to future innovations.

 

  1. What does IBM foresee will change in the landscape of cognitive computing?
  • We foresee that cognitive computing in the coming years will continue to make huge advancements in technology that will make our systems even smarter.
  • Through continuous human interactions, these systems will be able to learn at a much greater rate and continuously learn up to the point where they can fully run without human intervention.
  • As cognitive systems continue to be refined, we can also expect these systems to tackle an even broader range of tasks that even humans may not be able to do.
[adrotate group=”2″]
  1. How has IBM globally performed financially in the past year of 2016?
  • 2016 was another solid year for IBM as we have achieved several milestones in the year.
  • Aside from working some of the world’s largest most well-known companies across industries, we have also invested into our capabilities such as cloud, analytics and cognitive that has now accounted for 41% of our total revenue.
  • We have also not lost sight of our goal of developing innovations in 2016 with over 8,000 patents registered in 2016 and we are currently leading in terms of U.S. patents earned for the 24th year in a row.

 

  1. What are some of the upcoming plans that IBM have in the year of 2017?
  • In the coming year, we are looking to continue our research into our field of expertise for designing and developing innovative new systems that can enhance our client’s business and help them succeed in their field.
  • We have some projects in Malaysia in the banking, manufacturing and healthcare industries.

Go Back To > First PageEvents | Home

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!