Page 243 - Ai_C10_Flipbook
P. 243
Different Types of Chatbots
After interacting with different types of chatbots you must have understood that there are some chatbots that
follow some predefined rules to hold the conversation between the user and the machine. While, some are made
intelligent as they can handle patterns and common human behaviours while others are smart AI machines which
can work, think and learn from their experience like a human. Chatbots are generally categorised into two types:
Script-bot and Smart-bot.
Script-bot
A Script-Bot is a simple chatbot with limited functionalities. These chatbots follow predefined rules and execute
specific tasks based on a scripted flow. Since they operate within a structured framework, they are easy to develop
and require little to no programming knowledge.
These chatbots are best suited for straightforward tasks such as, automating home functions (e.g., turning off
lights with a voice command) and ordering food online through basic menu selections. Some of the companies
use them as customer care services for providing interactive frequently asked questions services to the customers
and if these chatbots are unable to handle certain queries then they connect a company’s customer care employee
directly to the customer.
Smart-bot
Smart-bot are flexible, powerful, AI-based models that have wider functionalities and support machine learning
algorithms that make a machine learn from the experience. They simulate human-like interactions with the users.
These chatbots require complex programming and work on bigger databases and other resources to help the
model understand the context of interactions.
Different types of virtual assistants like Google Assistant, Amazon Alexa, Siri, etc. are best examples of Smart bots.
The difference between Script-bot and Smart-bot are as follows:
Feature Script-bot Smart-bot
Definition A rule-based bot that follows predefined AI-driven bot capable of learning, adapting,
scripts or workflows. and handling complex tasks.
Intelligence Level Low: Only performs tasks based on High: Can use AI and machine learning to
hardcoded instructions. make decisions.
Adaptability Fixed functionality; cannot adapt to new Can adapt to new scenarios and learn from
scenarios unless reprogrammed. interactions.
Response Handles simple, straightforward tasks Handles dynamic, context-aware, and nuanced
Complexity with limited scope. conversations.
Learning Ability No learning capability; requires manual Can improve through machine learning and
updates to change behaviour. feedback loops.
Flexibility Limited to specific tasks and workflows. Works across a variety of tasks and domains.
Examples Basic FAQ bots, data-entry bots, or ChatGPT, customer service AI, personal
script-following automation. assistants like Siri or Alexa.
Development Low to Medium: Requires scripting and High: Requires AI training, natural language
Complexity workflow definitions only. processing (NLP), and ongoing updates.
Text Processing
Making a computer understand natural language is a complex process. First, we need to understand that humans
interact using characters, words, and sentences, while machines interact using numbers. So, to make the machine
learn and process a sentence in terms of numbers we first need to follow a Pre-Processing stage of NLP about
which we will study in detail.
Natural Language Processing 241

