Chat bots are Conversational agents or a Dialog system that simulates Intelligent conversation using text or speech.
Why chat bots?
In recent times, a sudden rise in interest towards yet to be explored field of making smart bots with AI capabilities and conversational human-computer interaction as the main paradigm could be observed. Microsoft betting big on bots, Facebook drawing line for Messenger using bots, Google bringing intelligence in messaging using bot. N-number startup pop out with conversational commerce and business.
A study shows 90% of our online time is spent on emails and messaging platforms. Another study revealed internet usage on mobile devices is greater than on desktops.
“Business happens where people are.”
How Bot works?
In order to build a good conversational interface, we need to look beyond a simple search by a substring or regular expressions that we usually use while dealing with strings.
The task of understanding spoken language and free text conversation in plain English is not as straightforward as it might seem from the first look.
AI has always been a land of exploration.
Models that are most often used for Smart conversational bots –
Retrieval-based models (easier)
Glimpse of how this model works : https://www.hallwaze.com/hallwaze/u/250k
It is based on pre-defined responses and some heuristic search (whether be rule-based expression match or fancy complex machine learning classifiers).
They never generate new text but pick perfect response.
Grammatically correct and appropriate answers.
Generative models (harder)
Example on how this generative model works – https://www.hallwaze.com/hallwaze/u/250l
Doesn’t rely on pre-defined responses – generates new responses from scratch.
Generative models are typically based on Machine Translation techniques. Input to Output translation using vector.
Low chance of being grammatically correct and appropriate.
It’s easy to train machines and prepare a system, because of known perspective. As the scope of problem being dealt with is limited, possible inputs and outputs are also somewhat limited while the system tries to achieve a very specific goal. For Example – Technical Customer Support or Shopping Assistants
Open Domain –
Context could be anything and is volatile in nature. Large data set is needed to train generative models and yet accuracy remains a major challenge.
Length of conversation –
Longer the conversation, harder it becomes to keep a track of what is said and when. Constant change in context is another big challenge to be dealt with.
To be continued in part 2!
- Implementing bot in IoT ecosystem
What we did... With the help of Hallwaze beat’s RTM api and a bot engine that uses popular Wit.ai for NLU and NLP. We cooked some very interesting bots. We…
- Implementing Chatbot in IoT Ecosystem (Use-case)
In continuation with the Chatbot series, we are back with our next blog on implementing chatbot in IoT ecosystem. Let’s get started. What Did We Do? With the help of…
- AI in Banking(2) – Chatbots
Hello readers, as promised, we are here with our 2nd blog in the series of blogs on “AI in Banking and Financial Services”, focusing on Chatbots. As discussed in the…
- ChatBots (Part 2) – How to make Bots
“It is said that to explain is to explain away. This maxim is nowhere so well fulfilled as in the area of computer programming, especially in what is called heuristic…
- A Know-how of Timesheet Management via Chatbot
SAP Timesheet Management using Chatbot is an application, which helps users in managing their time entries quickly and efficiently in real-time through mobile phones. Chatbot is an AI-powered computer program…
- Mobile App Automation Testing using ‘ESPRESSO’
If you are a Mobile Apps Test Engineer, you cannot overlook the very reliable Google Product i.e. Espresso. Espresso is an automatic UI testing or as we call it “hands…