Building intelligent chat bots : An AI dream

The Artificial intelligence market is expected to reach $5.05 billion by 2020.The global chat bots market was valued at USD 88.5 Million in 2015 and is anticipated to witness a substantial compound annual growth rate (CAGR) of 35.08% over the period 2016-2023. Chat bots don’t have the inherent ability to recognize specific human emotions – like irony, sarcasm etc.

Rise of chat bots needs to go hand in hand with built-in sentimental capabilities to enhance and add value to rising customer expectations and experiences. Thus chat bots need to be a disruptive sink of human sentiments, bridging the gap between customer service assistants and domain specific chat bots trained for customer evaluations. Also, the recent Indigo Airlines mix up, where a chat bot was unable to recognize human sentiments was badly rated on twitter and adds context to the problem we are trying to solve. This could save up to billions of dollars in industries which are and which will be driven by chat bots in the near future. In this context, the total addressable market for retail, travel, hospitality and consumer goods is 1.5 trillion $.

Trends driving chat bot usage in current and future markets include mobile messenger domination, app fatigue, support for chat bots by Facebook, Microsoft and other leaders, dramatic reduction in chat bot development costs. Also, advanced automation will move towards AI, deep learning, internet of things (IoT) and smart devices, cognitive computing, machine learning and sentiment analysis by 2020.

The customer experience is built through the whole life cycle of the product or service as well as through the customer’s own experience. Brand equity is nurtured through various channels and especially over internet. Moreover, social media is now the game changer for any business to boom. Omni channel is becoming a reality with multiple hybrid scenarios: 

? Order on line, go and pick up

? Visit and get delivered

? Visit and order on line for home delivery

In the current scenario, without chat bots customers may face difficulties in getting a solution for certain queries. The time and agony concerned to undelivered and wrongly delivered services might increase customer anxiety and sarcasm which, again will create negative impact.

Building a chat bot using natural language processing will be the key to address opportunities in the market. The behavior and intent detection in chat bots has to be trained to express and recognize positive sentiments which can add positive value to a brand as well as detection of negative emotions has to be embraced with appropriate left shift thinking to deliver unique and compelling customer experiences. Hence, building sentience in chat bots will augment and enhance customer experiences. 

The customer will use this as follows:

1. Go to the chatbot URL and request a service. 

2. Leverage the chatbots natural language capabilities to request or deliver services. 

3. Chat with the chatbot.

4. Mine human sentiments to understand the quality of services opted for. 

5. Translate the experience to a points based system which adds reward points for consumers.

6. Enhance the usage to allow for customer growth & retention by driving traffic to the app via gamification indicators.

Research is being done to address sentiments like irony, sarcasm etc. SquadRun has time-tested this problem with their crowd sourcing solution. But, apparently – they have dealt with the problem only by using human intelligence.

Thus, the future of chat bots industry is heading towards smart and meaningful customer interactions and social insights. Future bots will augment and enhance, rather than automate and replace, our human experiences.

A computer would deserve to be called intelligent if it could deceive a human into believing it was human. : Alan Turing