Can Artificial Intelligence make a difference or are we talking cows flying across the blue sky?

by November 3, 2016 0 comments

Contributed by Sauvik Banerjjee, CTA of SAP global

The perceptions surrounding artificial intelligence and its implications in the eye of the public have lead to expectations like that from a magician! Everything can get done without any effort. This brings me to two very important questions. One is what can artificial intelligence do for us, specifically for India in the next 2 decades – give some change that we can see in our lifetimes. The other question is what does India need in the next 2 decades from artificial intelligence to make a tangible difference to the lives of people here.

To answer the second question, I think it is imperative to answer the first most pragmatically. And to do that efficiently, we need to focus on what are the key aspects of artificial intelligence. Simplistically speaking, it comprises of:

  • areas of natural language processing
  • deep learning through various neural network modeling
  • algorithm based personalization
  • chats
  • recommendation modules

All of these finally culminate into a marriage between intelligence of softwares augmenting the intelligence of machines.

To take it to the next level, let’s explore the example of the now popular Terminator Age. Prima Facie the conversation revolves around the potential impact of a machine learning process driven by software and neural modeling handling gazillion data packets. Now what use are data packets? Where do they take us? Why do this entire thing at all?

A simple use case, demonstrates the potential. Over 800 million Indians do not even have access to hospitals and clinics with sufficient hardware. The impact is not just in their healthcare opportunities but also the loss of extraordinary amounts of data – knowledge – that in times of epidemics could prove to be game changers for the country.

pic-2

Sauvik Banerjee, CTA, SAP Global

Critical data on symptoms, diseases and chronic conditions accumulate continuously and neither are being collected nor used constructively to lead to any potential short term benefit – leave alone the long term benefits. Consider this use case: A simple machine learning algorithm can analyze and predict and recommend the MRI scanners while the scan of a patient is going on! This will land up fast tracking the process of remedies the doctors or the clinics can recommend. In times of crisis – this simple measure can be the difference between life and death of a patient!

What therefore should be expected from AI?

A software with a decision tree model is sending the data to the machine to show case all capabilities for symptoms to remedies. The whole life cycle. Now imagine that the next generation of MRI scanners, radiology departments and Health care clinics of private and public sector has access to this piece of technology as a package. We might be looking at killing of chronic diseases amidst our tier 1, 2 and the villages where the process of diagnosis is real time and fast tracked with remedies. Computers can read and learn similar outbreaks anywhere across the country giving doctors similar histories, the actual challenges to be expected and the medical route with the best chance of success. In case of an epidemic potentially arising, it helps the doctors and thus the government to control or minimize the impact or with diligent follow processes even limit the outbreak basis geographies.

This is not a utopian thought. It’s a reality that is just around the corner for many geographies!

Let’s now add to the level of complexity.

Consider the example of deep learning and the decision tree model and couple it with a NLP model and see the use case to the common man of India in the small towns and villages. Kirana stores in the remotes tier 2 town or village now are smart phone equipped. The village panchayets are gearing up for eGovernance. What will strengthen their faith will be a simple community assisted platform for detection of crop deficiency and crop protection.

Most farmers suffer from the ailment of not knowing when to protect his crop or what to do to protect his crop without the anytime access too chemicals. What if the model of a personal assistant has that information and in almost near realtime can get that information to the farmer with knowhow of what is the crop remedy, chemicals and recommendations by soil and crop experts across the country. How about taking it the next step and augmenting those algorithm based community platform via a soil sensory data mechanics to send it to the platform where deep learning capability can be powered by mechanics of IoT.

Hard to believe? But this again is not utopian! With dedicated private and public sector organisations working aggressively on this, it’s just the bat of an eyelid away. However, if you think that this is becoming crazy, we’ve not yet got to the point of sending cows fly through the clear blue skies!

Now lets take the popular use cases of the consumer world of online and offline commerce and break it down to chat bots and advanced personalization. Todays applications and India’s market places are all working in a secured mode to help and support their consumer base and CSR’s to ensure that quality of data being recommended is of business value. Yes they are trying to automate and ensure automated chat assistants, 3D modeling recommendations based on visual search recommendations based on neural network modeling- the use case is simple. I intend to purchase a show- I should be automatically prompted based on my user behavior the personalized closest patterns and matches of the product, through the assisted automated chat session I should be promoted in an interactive mode about their fulfillment- a social listening algorithm based on my social profile to start recommending the type for my shoe on my social platforms through contextual and push marketing but not only software driven. The advanced visual search algorithms, the automated assisted fulfillment will not only get my loyalty and trust back in a disloyal loss of trust mindset.

All of these interpretations of artificial intelligence aim at societal and economic problem solving. Digital India and connected India where value of human life and quality of day to day mechanics are made simple with artificial intelligence algorithms, technics and modeling. Artificial intelligence still remains to be a strong enabler to make human civilization smarter. Well Artificial intelligence applications in movies can perhaps also make cows fly across the clear blue skies, but really do we need it to do something like that or focus on the real need of real people in real life?

No Comments so far

Jump into a conversation

No Comments Yet!

You can be the one to start a conversation.

Your data will be safe!Your e-mail address will not be published. Also other data will not be shared with third person.