Information

Cognitive Technologies – Advancements in Cog Tech shaping your FUTURE!

Brain-like computing, human reasoning mechanisms, swarm intelligence were some of the hottest topics at the Ninth International Conference on Advanced Cognitive Technologies this year. To a non-specialist, these terms may seem to have dropped from a science fiction script, but these were just some of the rather arcane topics that were up for discussion at this conference.

Let’s drop things a notch (or maybe a bit more) and start at the basics so that we can begin to understand what these terms mean.

We start with looking into the term “Cognitive Technologies”:

Cognitive technologies, as “cognitive” suggests, refer to technologies that integrate the domains of Artificial Intelligence, Machine Learning, Pattern Recognition and Natural Language Processing to enable computers to emulate the way a human brain thinks and works.

Solutions created through Cognitive technologies have the ability to solve complex problems in a myriad variety of domains, as opposed to the vast bulk of current  AI solutions which provide solutions for specific domains. Such solutions have the potential to control themselves completely autonomously, with their own thoughts, worries, feelings, strengths, weaknesses, and predispositions.

Potential solutions include the augmenting of actual human brains, enabling individuals who have damaged brains to restore functionality akin to that of a normal person and much more.

The possibilities seem endless and are just being explored. The use of cognitive technologies in a business context is what we at WeCP find to be of interest.

Cognitive technologies in business:

The  business impacts of cognitive technologies can be rather interesting to forecast. The proliferation of such technologies has already started in some industries as we will see.

There are two broad ways in which ‘cogtech’ can benefit a generic business:

  1. Enable the users (here referring to the employees of the business) to enhance their current capabilities in their day to day work activities. Tasks that earlier required 100 working hours rare seen to be cut down to a fraction of the same.
  2. Enables the business to expand their operations without significantly expanding their human resources. There is a two-fold benefit to this as this leads to higher motivation levels of the employees who feel more accomplished and also leads to bottom line revenue increases which satisfy the business stakeholders’ goals as well.

The implementation of the technologies itself (in a business context) can be classified broadly into three parts:

  • Product
  • Process
  • Insight.

Product implementations refer to the products or services that an organization offers to its end customers being infused with solutions created with cognitive technology. This automatically enables the organization to provide better products or solutions to their end customers. This may be in terms of ease of use, efficiency, convenience, integration with existing solutions, etc.

Process implementations, however, refer to when the operational aspects of the organization are enhanced through the enhancement of the workflows used in the business. This may be accomplished through the use of automation to speed up menial tasks,  to the use of speech recognition and voice synthesis to take customer service calls.

Insight solutions use certain technologies like Machine Learning and Pattern Recognition as a base for the creation of incredibly powerful analytical tools and systems that will aid strategic decision making in an organization. This can range from determining the direction of marketing campaigns to the influencing of product development goals to keep them in-line with customer expectations.

 

WeCP’s cognitive technology assessments, and how they would help organizations:

The impact of cognitive technologies on organizations is growing, and the market for cognitive solutions is expected to increase at a 55.1% annual compound growth rate in 2016-20. The immense benefits that such technologies bring would no longer just be limited to certain parts of select organizations. In order to remain competitive, organizations are finding that integrating these technologies into their current operations has become paramount. The need for acquiring individuals who are erudite in these esoteric arts has become a considerable priority.

As such, publicly available research on these technologies is limited. Even more limited are the individuals who can call themselves experts in these areas. How would an organization assess the expertise of potential employees who would be handling these areas?

This is where we, at WeCP, come in with our assessments in cognitive technologies. We understand that only experts and entities possessing the intelligence and abilities of experts can assess other purported experts in an area like cognitive technology.

Our assessments on cognitive technologies are designed and created by a team of individuals who are themselves talented in this field. In addition to this, we leverage  WeCP’s signature concept of collective intelligence to ensure that experts from the fields of AI, ML, NLP and all the other myriad of domains that contribute to the overall cognitive technology mind-space are involved in the creation of the assessments, ensuring complete, comprehensive and exhaustive problems that are at a the cutting edge, as expected of assessments that assess the cream of the crop.

We provide here a couple of sample problems : 

Sample Problem 1. Your team has developed a website. When you launch the website on the internet, you provided a “comments” section for visitors to comment about the website. As soon as it was launched, many comments popped up within few minutes. You suspect that all of these were made by robots. Which of the given below techniques can you use to distiguish robots actual humans?

A) Enable authentication system for putting comments and use Captcha along with user credentianls for authenticating.

B) Enable biometric authentication system for putting comments.

C) Enable authetication system and use One Time Password along with authentication.

 

The motive behind asking this question is to access the knowledge of the examinee about the underlying implementation of Machine Learning. Also, the examinee is tested on the knowledge of the real life usage of Machine Learning, for example, Bot Detection. The examinee must be able to differentiate between the situations in which a machine can learn from patterns ( e.g. Captcha, Biometric security ) from the ones in which it cannot( e.g. OTP Security ).

 

Sample Problem 2.

Consider the following Logistic Regression model. 

Here, A, B and C are the inputs and D and E are the outputs. 

What is the output of D and E?

 

The motive behind asking this question is to access the knowledge of the candidate about the basic understanding of logistic regression model in machine learning.

 

Sample Problem 3.

Suppose you are given the following sentence:

“Losses are topping all calculations”

You are also told that

“Losses” has 2 possible tags: N and V

“are” has 1 possible tag: V

“topping” has 3 possible tags: N, ADJ, and V

“all” has 3 possible tags: N, ADV, and DT

“calculations” has 2 possible tags: N and V

(where N – Noun, V – Verb, ADJ – Adjective, ADV – Adverb, DT – Determiner)

How many tag sequences are possible for this sentence?

 

The motive behind this problem is to access the knowledge of the candidate about the applications of machine learning in parts-of-speech tagging problems in relation with natural processing language.

Leave a Reply

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