Chatbots - The Evolution of Today's Thinking Machines
- AI v Human Brain
- Artificial Intelligence
- big data
- chatbots definition
- Data Science
- Deep Learning
- Future of Life org
- IoT architecture
- IoT evolution
- IoT platforms
- Learning Centre
- Life 3.0
- Machine Learning
- smart healthcare
Exploring today's chatbot creations and where it all started.
What is a Chatbot? - Definition
A chatbot is a computer programme designed to simulate how a human would converse in a discussion. Also referred to as chatterbots and bots.
The simpler chatbot programmes scan the input (whether that be via audio or text) for keywords and then select the reply with the most matching keywords from a database. Some chatterbots use sophisticated natural language processing systems.
Alan Turing speculated that machines would be able to match human intellectual performance on certain tasks and suggested starting off “training” with an abstract task such as playing chess and secondly trying to find a way to simulate human cognitive development.
The evolution of AI did result in the development of chess playing computers which can outplay most amateurs and in the case of IBM’s Deep Blue, controversially defeated the world champion Garry Kasparov in 1997. (check out the documentary film on this The Man vs. The Machine).
Notable Chatbots – the early years
1966 Eliza by Joseph Weizenbaum - A simulation of a Rogerian therapist
1972 PARRY by Kenneth Colby - attempted to simulate a a paranoid schizophrenic
1995 ALICE by Richard Wallace - natural language processing chatterbot
1997 Jabberwacky by Rollo Carpenter - aimed to "simulate natural human chat in an interesting, entertaining and humorous manner".
Chatbots – the progress
2006 Watson by IBM - A question answering system that won the Jeopardy! contest, defeating the best human players in February 2011
2016 TAY by Microsoft - An artificial intelligence chatterbot that caused controversy on Twitter by releasing inflammatory tweets and was taken offline shortly after.
Chatbot creation - the ecosystem
The chatbot sector is experiencing massive growth. Since Facebook opened up the Messenger platform to developers last year, more than 100,000 unique chatterbots or chat agents have been developed. Specialist companies create chatbots for the fastest adopting sectors such as the travel industry, such as Yalochat and Digital Genius.
Nowadays you can even create your own bot using a chatbot creation website such as the one by Snatchbot.
As with other emerging technologies the chatbot sector is producing micro-specialisms around the different capabilities of chatbots to “communicate and converse”. This includes the likes of sentiment analysis where the bot can reportedly recognise emotion from facial expressions, eye contact and tone of voice and bots they can express emotion and appear more humanlike.
Many chatbots now include functional features such as web searching abilities.