Chatbots

Chatbots are AI systems that can converse with humans using natural language. They range from simple rule-based systems to sophisticated LLM-powered assistants.

12 min•By Priygop Team•Updated 2026

Types of Chatbots

Rule-based chatbots: follow a script of predefined responses. If the user says X, respond with Y. Simple, predictable, but cannot handle questions outside the script.

Retrieval-based chatbots: search a database of pre-written answers and pick the best match for the user's question.

Generative chatbots: use a language model to generate responses from scratch. ChatGPT is a generative chatbot. More flexible but can make mistakes.

Diagram

Loading diagram…

Deep Learning ⊂ Machine Learning ⊂ Artificial Intelligence

Simple Rule-Based Chatbot

# Simple rule-based customer support chatbot

class SimpleChatbot:
    def __init__(self, name="Assistant"):
        self.name = name
        self.rules = {
            # keywords: response
            "hello|hi|hey": f"Hello! I am {name}. How can I help you today?",
            "order|track|shipping": "To track your order, visit our website and click 'My Orders'. Enter your order number to see the status.",
            "return|refund": "We offer free returns within 30 days. Please go to our website and click 'Start a Return' to begin the process.",
            "hours|open|close": "Our customer service team is available Monday to Friday, 9am to 6pm.",
            "price|cost|how much": "You can find current prices on our website. Use code SAVE10 for 10% off your first order.",
            "bye|goodbye|thanks|thank": "Thank you for contacting us! Have a great day.",
        }
        self.default = "I am not sure I understand. Could you rephrase that, or would you like to speak with a human agent?"

    def respond(self, user_input):
        user_lower = user_input.lower()
        for pattern, response in self.rules.items():
            keywords = pattern.split("|")
            if any(kw in user_lower for kw in keywords):
                return response
        return self.default


# Simulate a conversation
bot = SimpleChatbot("ShopBot")
conversation = [
    "Hi there!",
    "I want to know about returning a product",
    "What are your opening hours?",
    "How much does shipping cost?",
    "Thanks, goodbye",
]

print("Chatbot Conversation:")
print()
for user_msg in conversation:
    print(f"  You: {user_msg}")
    response = bot.respond(user_msg)
    print(f"  {bot.name}: {response}")
    print()

Try It Yourself

Try It YourselfHTML

HTML Editor

✓ ValidTab = 2 spaces

class SimpleChatbot:
    def __init__(self):
        self.rules = {
            "hello|hi": "Hello! How can I help you?",
            "name": "I am a simple AI chatbot.",
            "weather": "I cannot check the weather, but I recommend a weather app!",
            "joke": "Why do programmers prefer dark mode? Because light attracts bugs!",
            "bye|goodbye": "Goodbye! Have a wonderful day.",
        }
    
    def respond(self, user_input):
        user_lower = user_input.lower()
        for pattern, response in self.rules.items():
            if any(kw in user_lower for kw in pattern.split("|")):
                return response
        return "I don't understand. Try: hello, name, weather, joke, or bye."

bot = SimpleChatbot()

# Try different messages:
messages = ["hi", "what is your name", "tell me a joke", "goodbye"]
for msg in messages:
    print(f"You: {msg}")
    print(f"Bot: {bot.respond(msg)}")
    print()

Result

Click ▶ Run to see the result

Edit the code on the left, then click Run (Ctrl+Enter)

HTML|25 lines|939 chars|✓ Valid syntax

UTF-8

Topics in This Module