Guest Article: Flourishing Cognitive Systems - HPC ASIA
Features

Guest Article: Flourishing Cognitive Systems

CostConsulting
pic

Sushil Pramanick

Cognitive computing dates back to the late 1960s when initial innovation was applied to supercomputers. However, decision-making capability, simulation, and self-learning intelligence were limited. Throughout its evolution, cognitive computing researchers have made significant strides because of major advances in cognitive behavior science, computational intelligence, and computing power.

Particularly in the fields of advanced medical science and neuropsychology, computer scientists were able to study the mechanics of the human brain, which allowed scientists to build computational simulations that modeled the human mind. These models enabled various attributes and parameters from past experiences to be associated into cognitive systems. As a result, cognitive computing and decision science progressed, allowing scientists to develop computers that operated at a higher rate of speed and accuracy than the human brain was capable of processing.

At IBM, cognitive computing systems learn and interact naturally with people to extend what either humans or machines are capable of doing on their own. They help human experts make better-informed decisions by penetrating the complexity of big data. The growth of big data is accelerating as increasing amounts of the world’s activity is expressed in digital form. Not only is this data increasing in volume, but it is also growing in velocity, variety, and uncertainty.

A majority of data now comes in unstructured forms such as video, images, symbols, and natural language—which calls for an advanced computing model that businesses need to process and make sense of the data. The computing model also enhances and extends the expertise of humans. Rather than being programmed to anticipate every possible answer or action needed to perform a function or set of tasks, cognitive systems are trained by using artificial intelligence (AI) and machine-learning algorithms to infer, predict, sense, and—in some ways—think.

Elasticity

Cognitive computing reassesses the nature of relationships between multiple variables and environment factors. It helps evaluate the nature of the relationship between people and the increasingly pervasive digital environment they find themselves in today. Cognitive computing systems may play the role of assistant or coach for the end user, or they may act virtually autonomously in many problem-solving situations.

The boundaries of the processes and domains these systems will affect are still elastic and emergent. These systems must learn as attributes and variables change, even as questions and decisions evolve. They must also resolve ambiguity and tolerate unpredictability, and they need to be engineered to feed on dynamic data in real time or near–real time.

In addition, cognitive computing systems should interact easily with end users so that they can define their needs with clear specifications and pseudo algorithms, and they may also interact with other processors, devices, and cloud services, as well as with people. By asking questions or finding additional source input if a problem statement is ambiguous or incomplete, cognitive computing systems provide assistance in defining a problem. And using the capability to remember previous interactions in a process, these systems can return information that is suitable for a specific application at a particular point in time.

Moreover, cognitive computing systems must understand, identify, and extract contextual elements such as appropriate domain, end-user profile, goal, location, meaning, process, regulations, syntax, task, and time. They may draw on multiple sources of information, including both structured and unstructured digital information and sensory inputs—auditory, gestural, sensor provided, or visual. The output from cognitive systems may be instructive, prescriptive, suggestive, or simply entertaining.

Extensibility

Cognitive informatics is a leading-edge and multidisciplinary research field that tackles the fundamental problems shared by AI, cognitive science, computation, computational intelligence, cybernetics, economics, life sciences, linguistics, management science, medical science, modern informatics, neuropsychology, philosophy, software engineering, and systems science. The development and cross-fertilization between these science and engineering disciplines have led to a wide range of emerging research areas known as cognitive informatics. Several areas offer example use cases for cognitive computing.

Banking

In fraud detection, banking institutions can apply cognitive tools to go beyond analyses of cardholders’ credit transaction histories. Cognitive computing may provide these organizations with advanced associational intelligence, such as when individuals are most likely to make purchases, what they are likely to buy, and under what circumstances they make the purchase.

Finance

Financial advisors are also expected to benefit from cognitive computing. This user group includes individuals who handle their own portfolios, as the technology enables bringing together current, personalized, and relevant information and remembering questions and preferences.

Medical

Evidence-based learning, hypothesis generation, and natural-language skills can help medical professionals make key decisions in patient diagnosis and treatment. The objective is to give doctors and surgeons a quick way to access diagnostic and treatment options based on updated research.

Shopping

The capability of cognitive computing systems to evaluate and generate hypotheses is expected to help retail industries find correlations, insights, and patterns in mountains of unstructured and structured data. The IBM® Watson™ environment’s app development platform is already moving into this physical-virtual space. For example, Fluid, a start-up, has layered Watson on top of its Expert Personal Shopper app for retail brands. Watson can be a buyer’s personal shopping assistant. Store associates may also have similar intelligent technology providing them instant customer loyalty data, end-user reviews and blogs, magazine articles, product information, and sales histories. As a result, when shoppers do need to talk with another human, sales associates know exactly how they can help.

Weather forecasting and planning

Weather forecasting can benefit from cognitive computing and big data analytics. For example, the IBM Deep Thunder™ service, a research project that creates precise, local weather forecasts, can predict severe storms in a specific area up to three days before the event. This early-warning system gives local authorities and residents enough time to make preparations.

Practicality

According to Deloitte University Press as stated in their Cognitive Analytics published in Feb 20141, cognitive analytics helps address some key challenges. It can improve prediction accuracy, provide augmentation and scale to human cognition, and allow tasks to be performed more efficiently—and automatically—through context-based suggestions. For organizations that want to enhance their ability to sense and respond, cognitive analytics offers a powerful way to bridge the gap between the promise of big data and the reality of practical decision making.

IBM has calculated that the market size for cognitive computing services in the private sector is in the neighborhood of USD50 billion2.At present, there are very few vendors in the field. While IBM has announced the creation of the Watson Group to commercialize cognitive computing and Google has acquired an AI start-up, DeepMind, there are few companies working in this space. Much of the work is still happening at a university level or within research organizations.

Cognitive computing is still early from a commercialization perspective. An industry-wide adoption, and its impact on a wide range of organizations, is likely another three-to-five years off. For a while at least, cognitive computing is expected to fill a niche in industries such as healthcare and financial markets, where highly complex decision making is the norm. Eventually, cognitive computing is intended to become a standard tool in every corporate toolbox, helping to augment human decision making.

References:
1. Cognitive Analytics – by Deloitte University Press

http://dupress.com/articles/2014-tech-trends-cognitive-analytics/

2. IBM’s cognitive computing: The next wave applications that are as intutive & capable as human beings

http://articles.economictimes.indiatimes.com/2013-12-10/news/45035353_1_ibm-cognitive-computing-watson-solutions

Author BIO: 

Sushil Pramanick is an Associate Partner at IBM Global Business Services and a senior information technology executive. Sushil has consistently ranked in Top 50/100/200 industry rankings in Big Data Analytics and Data Science. Sushil is also founder and president of a global non-profit origanization – The Big Data Institute (TBDI).

 



  

Subscribe to our mailing list

 

* indicates required

 

 

 

 

 

 

 

 

 

 

 Newsletter Frequency

 

 

 

 

 

 Email Format

 

 

 

 

 

   

 

 

 

 


Comments

comments

Click to comment

Leave a Reply

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

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

To Top