I am a marketing intern at Valuefirst Digital Media. I write blogs on AI, Machine Learning, Chatbots, Automation etc for House of Bots. ...

Full Bio 
Follow on

I am a marketing intern at Valuefirst Digital Media. I write blogs on AI, Machine Learning, Chatbots, Automation etc for House of Bots.

Why is there so much buzz around Predictive Analytics?
1033 days ago

Changing Scenario of Automation over the years
1034 days ago

Top 7 trending technologies in 2018
1035 days ago

A Beginner's Manual to Data Science & Data Analytics
1035 days ago

Artificial Intelligence: A big boon for recruitment?
1036 days ago

Top 5 chatbot platforms in India

Artificial Intelligence: Real-World Applications

How can Big Data Analytics influence Decision Making?

Why is there so much buzz around Predictive Analytics?

Levels of Big Data Maturity

Progress Teaches Chatbots To 'Talk'

By shiwaneeg |Email | Mar 21, 2018 | 9444 Views

Artificial Intelligence (AI) is in everything. To be clear, AI has now become a part of the way every layer and thread of the total IT fabric is being engineered. At the 'chipset' processor level, we find AI being used to aid hardware engineering. Upwards from the motherboard we also now find AI being used perhaps more obviously in our software applications. Upwards further still, we can now look to network management as a higher level area for AI to applied to our IT structures.

In the midst of this AI popularization (for want of a less denigrating term) technology vendors are now trying to apply a degree of machine brain understanding (or machine learning, as we tend to say) into every aspect of the apps we use every day.

Chatty robots

Thinking back to science fiction movie style notions of AI, the original concepts centered around computers being able to 'talk' to us and understand what we are saying. In contemporary terms, this is the stuff of speech recognition, Natural Language Processing (NLP) and contextualized software programs that know how to interact with us. Not robots, but 'chatbots' â?? that is, dedicated software programs built to provide conversational interactions with humans using a pre-programmed level of vocabulary and understanding.

Self-styled 'cognitive-first applications' company Progress is now attempting to put new layers of AI into the chatbot experience. The company's eponymously named Progress NativeChat product is an AI-driven platform for creating and deploying chatbots. The software uses a branded tool known as CognitiveFlow to help train chatbots with goals, examples and data from existing backend company systems, similar to the process for training new customer service agents.

The chatbot needs brain DNA in order to understand some of the world around it and the way people act and the kinds of things that they ask for â?? this is that DNA.

"Chatbots solve many pressing customer service challenges for organizations - freeing overloaded call center employees to focus on critical business needs by offloading transactional requests to increase customer satisfaction. Chatbots are most powerful when they are intelligent and context-aware, directly pulling data and insights from systems of record," said Dmitri Tcherevik, CTO Progress.

Tcherevik insists that NativeChat makes it easy to create chatbots on top of existing systems that interact with users in a natural way and improve contextual understanding, accuracy and forecasting in conversations. NativeChat can be integrated into self-service web portals and mobile apps to create what Progress is calling customer self-service across any channel. Customers can also communicate using NativeChat through social channels such as Facebook messenger and other live chat technologies. NativeChat also integrates with any enterprise system that supports REST APIs.

Building robot brains

NativeChat is the first product to come out of Progress Labs, the innovation incubator inside of Progress. The product can support both transactional and FAQ-style interactions and perform natural conversation understanding in 72 human languages. We can now use software like this to train chatbots from existing FAQ pages and other written materials.

The interesting part is what happens next i.e. we are already talking about the need to make Artificial Intelligence both neutral of prejudice but, at the same time, also cognizant and aware of cultural considerations.

As chatbots now start to become more human, they will mostly need to be gender neutral, able to deal with ambiguities and intelligent enough to be able to 'extract intent' from the human who is attempting to talk to them. Progress provides a list of best practices for building a chatbot brain and reminds us that we need to establish limits so that we know what our software brains do not know. Chatbots can be annoying, but perhaps we're on the cusp of understanding how to make them 'talk' to us better.

This article was originally published in Forbes

Source: Forbes