You've learned this already. ✅
Click here to view the next lesson.
Chapter 10: Introduction to Chatbots
Practical Exercises
Exercise 1: Rule-Based Chatbot
Task: Implement a rule-based chatbot that can answer basic questions about a fictional company's contact information, hours of operation, and services offered.
Solution:
def rule_based_chatbot(user_input):
responses = {
"hello": "Hello! How can I assist you today?",
"hi": "Hi there! What can I do for you?",
"what are your hours of operation?": "Our office is open from 9 AM to 5 PM, Monday to Friday.",
"where are you located?": "We are located at 123 Main Street, Anytown, USA.",
"what services do you offer?": "We offer web development, mobile app development, and digital marketing services.",
"bye": "Goodbye! Have a great day!"
}
user_input = user_input.lower()
return responses.get(user_input, "I'm sorry, I don't understand that. Can you please rephrase?")
# Test the chatbot
while True:
user_input = input("You: ")
if user_input.lower() == "exit":
print("ChatBot: Goodbye! Have a great day!")
break
response = rule_based_chatbot(user_input)
print(f"ChatBot: {response}")Exercise 2: Retrieval-Based Chatbot
Task: Implement a retrieval-based chatbot that uses TF-IDF vectorizer and cosine similarity to match user queries with predefined responses.
Solution:
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity
corpus = [
"Our office is open from 9 AM to 5 PM, Monday to Friday.",
"We are located at 123 Main Street, Anytown, USA.",
"We offer web development, mobile app development, and digital marketing services."
]
vectorizer = TfidfVectorizer()
X = vectorizer.fit_transform(corpus)
def retrieval_based_chatbot(user_input):
user_input_vector = vectorizer.transform([user_input])
similarities = cosine_similarity(user_input_vector, X)
response_index = similarities.argmax()
return corpus[response_index]
# Test the chatbot
while True:
user_input = input("You: ")
if user_input.lower() == "exit":
print("ChatBot: Goodbye! Have a great day!")
break
response = retrieval_based_chatbot(user_input)
print(f"ChatBot: {response}")Exercise 3: Generative Chatbot
Task: Implement a simple generative chatbot using an existing Seq2Seq model for generating responses. For this exercise, we'll simulate the use of a pre-trained model.
Solution:
Since training a Seq2Seq model from scratch is complex and requires a significant amount of data, we will simulate the responses as if generated by a pre-trained Seq2Seq model.
# Simulated pre-trained Seq2Seq model responses
pretrained_responses = {
"hello": "Hello! How can I help you today?",
"how are you?": "I'm just a chatbot, but I'm here to assist you!",
"what is your name?": "I am ChatBot, your virtual assistant.",
"bye": "Goodbye! Take care!"
}
def generative_chatbot(user_input):
user_input = user_input.lower()
return pretrained_responses.get(user_input, "I'm still learning. Can you please rephrase?")
# Test the chatbot
while True:
user_input = input("You: ")
if user_input.lower() == "exit":
print("ChatBot: Goodbye! Take care!")
break
response = generative_chatbot(user_input)
print(f"ChatBot: {response}")Exercise 4: Hybrid Chatbot
Task: Implement a hybrid chatbot that uses rule-based responses for common greetings and self-learning (retrieval-based) responses for other queries.
Solution:
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity
corpus = [
"Our office is open from 9 AM to 5 PM, Monday to Friday.",
"We are located at 123 Main Street, Anytown, USA.",
"We offer web development, mobile app development, and digital marketing services."
]
vectorizer = TfidfVectorizer()
X = vectorizer.fit_transform(corpus)
def hybrid_chatbot(user_input):
rule_based_responses = {
"hello": "Hello! How can I assist you today?",
"hi": "Hi there! What can I do for you?",
"bye": "Goodbye! Have a great day!"
}
user_input_lower = user_input.lower()
if user_input_lower in rule_based_responses:
return rule_based_responses[user_input_lower]
else:
user_input_vector = vectorizer.transform([user_input])
similarities = cosine_similarity(user_input_vector, X)
response_index = similarities.argmax()
return corpus[response_index]
# Test the chatbot
while True:
user_input = input("You: ")
if user_input.lower() == "exit":
print("ChatBot: Goodbye! Have a great day!")
break
response = hybrid_chatbot(user_input)
print(f"ChatBot: {response}")Exercise 5: Enhancing the Hybrid Chatbot with More Responses
Task: Enhance the hybrid chatbot by adding more rule-based responses and expanding the retrieval-based corpus.
Solution:
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity
corpus = [
"Our office is open from 9 AM to 5 PM, Monday to Friday.",
"We are located at 123 Main Street, Anytown, USA.",
"We offer web development, mobile app development, and digital marketing services.",
"You can contact us at support@company.com.",
"Our phone number is (123) 456-7890."
]
vectorizer = TfidfVectorizer()
X = vectorizer.fit_transform(corpus)
def enhanced_hybrid_chatbot(user_input):
rule_based_responses = {
"hello": "Hello! How can I assist you today?",
"hi": "Hi there! What can I do for you?",
"what is your phone number?": "Our phone number is (123) 456-7890.",
"what is your email address?": "You can contact us at support@company.com.",
"bye": "Goodbye! Have a great day!"
}
user_input_lower = user_input.lower()
if user_input_lower in rule_based_responses:
return rule_based_responses[user_input_lower]
else:
user_input_vector = vectorizer.transform([user_input])
similarities = cosine_similarity(user_input_vector, X)
response_index = similarities.argmax()
return corpus[response_index]
# Test the chatbot
while True:
user_input = input("You: ")
if user_input.lower() == "exit":
print("ChatBot: Goodbye! Have a great day!")
break
response = enhanced_hybrid_chatbot(user_input)
print(f"ChatBot: {response}")These exercises provide hands-on experience with different types of chatbots, reinforcing the concepts covered in this chapter.
Practical Exercises
Exercise 1: Rule-Based Chatbot
Task: Implement a rule-based chatbot that can answer basic questions about a fictional company's contact information, hours of operation, and services offered.
Solution:
def rule_based_chatbot(user_input):
responses = {
"hello": "Hello! How can I assist you today?",
"hi": "Hi there! What can I do for you?",
"what are your hours of operation?": "Our office is open from 9 AM to 5 PM, Monday to Friday.",
"where are you located?": "We are located at 123 Main Street, Anytown, USA.",
"what services do you offer?": "We offer web development, mobile app development, and digital marketing services.",
"bye": "Goodbye! Have a great day!"
}
user_input = user_input.lower()
return responses.get(user_input, "I'm sorry, I don't understand that. Can you please rephrase?")
# Test the chatbot
while True:
user_input = input("You: ")
if user_input.lower() == "exit":
print("ChatBot: Goodbye! Have a great day!")
break
response = rule_based_chatbot(user_input)
print(f"ChatBot: {response}")Exercise 2: Retrieval-Based Chatbot
Task: Implement a retrieval-based chatbot that uses TF-IDF vectorizer and cosine similarity to match user queries with predefined responses.
Solution:
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity
corpus = [
"Our office is open from 9 AM to 5 PM, Monday to Friday.",
"We are located at 123 Main Street, Anytown, USA.",
"We offer web development, mobile app development, and digital marketing services."
]
vectorizer = TfidfVectorizer()
X = vectorizer.fit_transform(corpus)
def retrieval_based_chatbot(user_input):
user_input_vector = vectorizer.transform([user_input])
similarities = cosine_similarity(user_input_vector, X)
response_index = similarities.argmax()
return corpus[response_index]
# Test the chatbot
while True:
user_input = input("You: ")
if user_input.lower() == "exit":
print("ChatBot: Goodbye! Have a great day!")
break
response = retrieval_based_chatbot(user_input)
print(f"ChatBot: {response}")Exercise 3: Generative Chatbot
Task: Implement a simple generative chatbot using an existing Seq2Seq model for generating responses. For this exercise, we'll simulate the use of a pre-trained model.
Solution:
Since training a Seq2Seq model from scratch is complex and requires a significant amount of data, we will simulate the responses as if generated by a pre-trained Seq2Seq model.
# Simulated pre-trained Seq2Seq model responses
pretrained_responses = {
"hello": "Hello! How can I help you today?",
"how are you?": "I'm just a chatbot, but I'm here to assist you!",
"what is your name?": "I am ChatBot, your virtual assistant.",
"bye": "Goodbye! Take care!"
}
def generative_chatbot(user_input):
user_input = user_input.lower()
return pretrained_responses.get(user_input, "I'm still learning. Can you please rephrase?")
# Test the chatbot
while True:
user_input = input("You: ")
if user_input.lower() == "exit":
print("ChatBot: Goodbye! Take care!")
break
response = generative_chatbot(user_input)
print(f"ChatBot: {response}")Exercise 4: Hybrid Chatbot
Task: Implement a hybrid chatbot that uses rule-based responses for common greetings and self-learning (retrieval-based) responses for other queries.
Solution:
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity
corpus = [
"Our office is open from 9 AM to 5 PM, Monday to Friday.",
"We are located at 123 Main Street, Anytown, USA.",
"We offer web development, mobile app development, and digital marketing services."
]
vectorizer = TfidfVectorizer()
X = vectorizer.fit_transform(corpus)
def hybrid_chatbot(user_input):
rule_based_responses = {
"hello": "Hello! How can I assist you today?",
"hi": "Hi there! What can I do for you?",
"bye": "Goodbye! Have a great day!"
}
user_input_lower = user_input.lower()
if user_input_lower in rule_based_responses:
return rule_based_responses[user_input_lower]
else:
user_input_vector = vectorizer.transform([user_input])
similarities = cosine_similarity(user_input_vector, X)
response_index = similarities.argmax()
return corpus[response_index]
# Test the chatbot
while True:
user_input = input("You: ")
if user_input.lower() == "exit":
print("ChatBot: Goodbye! Have a great day!")
break
response = hybrid_chatbot(user_input)
print(f"ChatBot: {response}")Exercise 5: Enhancing the Hybrid Chatbot with More Responses
Task: Enhance the hybrid chatbot by adding more rule-based responses and expanding the retrieval-based corpus.
Solution:
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity
corpus = [
"Our office is open from 9 AM to 5 PM, Monday to Friday.",
"We are located at 123 Main Street, Anytown, USA.",
"We offer web development, mobile app development, and digital marketing services.",
"You can contact us at support@company.com.",
"Our phone number is (123) 456-7890."
]
vectorizer = TfidfVectorizer()
X = vectorizer.fit_transform(corpus)
def enhanced_hybrid_chatbot(user_input):
rule_based_responses = {
"hello": "Hello! How can I assist you today?",
"hi": "Hi there! What can I do for you?",
"what is your phone number?": "Our phone number is (123) 456-7890.",
"what is your email address?": "You can contact us at support@company.com.",
"bye": "Goodbye! Have a great day!"
}
user_input_lower = user_input.lower()
if user_input_lower in rule_based_responses:
return rule_based_responses[user_input_lower]
else:
user_input_vector = vectorizer.transform([user_input])
similarities = cosine_similarity(user_input_vector, X)
response_index = similarities.argmax()
return corpus[response_index]
# Test the chatbot
while True:
user_input = input("You: ")
if user_input.lower() == "exit":
print("ChatBot: Goodbye! Have a great day!")
break
response = enhanced_hybrid_chatbot(user_input)
print(f"ChatBot: {response}")These exercises provide hands-on experience with different types of chatbots, reinforcing the concepts covered in this chapter.
Practical Exercises
Exercise 1: Rule-Based Chatbot
Task: Implement a rule-based chatbot that can answer basic questions about a fictional company's contact information, hours of operation, and services offered.
Solution:
def rule_based_chatbot(user_input):
responses = {
"hello": "Hello! How can I assist you today?",
"hi": "Hi there! What can I do for you?",
"what are your hours of operation?": "Our office is open from 9 AM to 5 PM, Monday to Friday.",
"where are you located?": "We are located at 123 Main Street, Anytown, USA.",
"what services do you offer?": "We offer web development, mobile app development, and digital marketing services.",
"bye": "Goodbye! Have a great day!"
}
user_input = user_input.lower()
return responses.get(user_input, "I'm sorry, I don't understand that. Can you please rephrase?")
# Test the chatbot
while True:
user_input = input("You: ")
if user_input.lower() == "exit":
print("ChatBot: Goodbye! Have a great day!")
break
response = rule_based_chatbot(user_input)
print(f"ChatBot: {response}")Exercise 2: Retrieval-Based Chatbot
Task: Implement a retrieval-based chatbot that uses TF-IDF vectorizer and cosine similarity to match user queries with predefined responses.
Solution:
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity
corpus = [
"Our office is open from 9 AM to 5 PM, Monday to Friday.",
"We are located at 123 Main Street, Anytown, USA.",
"We offer web development, mobile app development, and digital marketing services."
]
vectorizer = TfidfVectorizer()
X = vectorizer.fit_transform(corpus)
def retrieval_based_chatbot(user_input):
user_input_vector = vectorizer.transform([user_input])
similarities = cosine_similarity(user_input_vector, X)
response_index = similarities.argmax()
return corpus[response_index]
# Test the chatbot
while True:
user_input = input("You: ")
if user_input.lower() == "exit":
print("ChatBot: Goodbye! Have a great day!")
break
response = retrieval_based_chatbot(user_input)
print(f"ChatBot: {response}")Exercise 3: Generative Chatbot
Task: Implement a simple generative chatbot using an existing Seq2Seq model for generating responses. For this exercise, we'll simulate the use of a pre-trained model.
Solution:
Since training a Seq2Seq model from scratch is complex and requires a significant amount of data, we will simulate the responses as if generated by a pre-trained Seq2Seq model.
# Simulated pre-trained Seq2Seq model responses
pretrained_responses = {
"hello": "Hello! How can I help you today?",
"how are you?": "I'm just a chatbot, but I'm here to assist you!",
"what is your name?": "I am ChatBot, your virtual assistant.",
"bye": "Goodbye! Take care!"
}
def generative_chatbot(user_input):
user_input = user_input.lower()
return pretrained_responses.get(user_input, "I'm still learning. Can you please rephrase?")
# Test the chatbot
while True:
user_input = input("You: ")
if user_input.lower() == "exit":
print("ChatBot: Goodbye! Take care!")
break
response = generative_chatbot(user_input)
print(f"ChatBot: {response}")Exercise 4: Hybrid Chatbot
Task: Implement a hybrid chatbot that uses rule-based responses for common greetings and self-learning (retrieval-based) responses for other queries.
Solution:
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity
corpus = [
"Our office is open from 9 AM to 5 PM, Monday to Friday.",
"We are located at 123 Main Street, Anytown, USA.",
"We offer web development, mobile app development, and digital marketing services."
]
vectorizer = TfidfVectorizer()
X = vectorizer.fit_transform(corpus)
def hybrid_chatbot(user_input):
rule_based_responses = {
"hello": "Hello! How can I assist you today?",
"hi": "Hi there! What can I do for you?",
"bye": "Goodbye! Have a great day!"
}
user_input_lower = user_input.lower()
if user_input_lower in rule_based_responses:
return rule_based_responses[user_input_lower]
else:
user_input_vector = vectorizer.transform([user_input])
similarities = cosine_similarity(user_input_vector, X)
response_index = similarities.argmax()
return corpus[response_index]
# Test the chatbot
while True:
user_input = input("You: ")
if user_input.lower() == "exit":
print("ChatBot: Goodbye! Have a great day!")
break
response = hybrid_chatbot(user_input)
print(f"ChatBot: {response}")Exercise 5: Enhancing the Hybrid Chatbot with More Responses
Task: Enhance the hybrid chatbot by adding more rule-based responses and expanding the retrieval-based corpus.
Solution:
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity
corpus = [
"Our office is open from 9 AM to 5 PM, Monday to Friday.",
"We are located at 123 Main Street, Anytown, USA.",
"We offer web development, mobile app development, and digital marketing services.",
"You can contact us at support@company.com.",
"Our phone number is (123) 456-7890."
]
vectorizer = TfidfVectorizer()
X = vectorizer.fit_transform(corpus)
def enhanced_hybrid_chatbot(user_input):
rule_based_responses = {
"hello": "Hello! How can I assist you today?",
"hi": "Hi there! What can I do for you?",
"what is your phone number?": "Our phone number is (123) 456-7890.",
"what is your email address?": "You can contact us at support@company.com.",
"bye": "Goodbye! Have a great day!"
}
user_input_lower = user_input.lower()
if user_input_lower in rule_based_responses:
return rule_based_responses[user_input_lower]
else:
user_input_vector = vectorizer.transform([user_input])
similarities = cosine_similarity(user_input_vector, X)
response_index = similarities.argmax()
return corpus[response_index]
# Test the chatbot
while True:
user_input = input("You: ")
if user_input.lower() == "exit":
print("ChatBot: Goodbye! Have a great day!")
break
response = enhanced_hybrid_chatbot(user_input)
print(f"ChatBot: {response}")These exercises provide hands-on experience with different types of chatbots, reinforcing the concepts covered in this chapter.
Practical Exercises
Exercise 1: Rule-Based Chatbot
Task: Implement a rule-based chatbot that can answer basic questions about a fictional company's contact information, hours of operation, and services offered.
Solution:
def rule_based_chatbot(user_input):
responses = {
"hello": "Hello! How can I assist you today?",
"hi": "Hi there! What can I do for you?",
"what are your hours of operation?": "Our office is open from 9 AM to 5 PM, Monday to Friday.",
"where are you located?": "We are located at 123 Main Street, Anytown, USA.",
"what services do you offer?": "We offer web development, mobile app development, and digital marketing services.",
"bye": "Goodbye! Have a great day!"
}
user_input = user_input.lower()
return responses.get(user_input, "I'm sorry, I don't understand that. Can you please rephrase?")
# Test the chatbot
while True:
user_input = input("You: ")
if user_input.lower() == "exit":
print("ChatBot: Goodbye! Have a great day!")
break
response = rule_based_chatbot(user_input)
print(f"ChatBot: {response}")Exercise 2: Retrieval-Based Chatbot
Task: Implement a retrieval-based chatbot that uses TF-IDF vectorizer and cosine similarity to match user queries with predefined responses.
Solution:
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity
corpus = [
"Our office is open from 9 AM to 5 PM, Monday to Friday.",
"We are located at 123 Main Street, Anytown, USA.",
"We offer web development, mobile app development, and digital marketing services."
]
vectorizer = TfidfVectorizer()
X = vectorizer.fit_transform(corpus)
def retrieval_based_chatbot(user_input):
user_input_vector = vectorizer.transform([user_input])
similarities = cosine_similarity(user_input_vector, X)
response_index = similarities.argmax()
return corpus[response_index]
# Test the chatbot
while True:
user_input = input("You: ")
if user_input.lower() == "exit":
print("ChatBot: Goodbye! Have a great day!")
break
response = retrieval_based_chatbot(user_input)
print(f"ChatBot: {response}")Exercise 3: Generative Chatbot
Task: Implement a simple generative chatbot using an existing Seq2Seq model for generating responses. For this exercise, we'll simulate the use of a pre-trained model.
Solution:
Since training a Seq2Seq model from scratch is complex and requires a significant amount of data, we will simulate the responses as if generated by a pre-trained Seq2Seq model.
# Simulated pre-trained Seq2Seq model responses
pretrained_responses = {
"hello": "Hello! How can I help you today?",
"how are you?": "I'm just a chatbot, but I'm here to assist you!",
"what is your name?": "I am ChatBot, your virtual assistant.",
"bye": "Goodbye! Take care!"
}
def generative_chatbot(user_input):
user_input = user_input.lower()
return pretrained_responses.get(user_input, "I'm still learning. Can you please rephrase?")
# Test the chatbot
while True:
user_input = input("You: ")
if user_input.lower() == "exit":
print("ChatBot: Goodbye! Take care!")
break
response = generative_chatbot(user_input)
print(f"ChatBot: {response}")Exercise 4: Hybrid Chatbot
Task: Implement a hybrid chatbot that uses rule-based responses for common greetings and self-learning (retrieval-based) responses for other queries.
Solution:
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity
corpus = [
"Our office is open from 9 AM to 5 PM, Monday to Friday.",
"We are located at 123 Main Street, Anytown, USA.",
"We offer web development, mobile app development, and digital marketing services."
]
vectorizer = TfidfVectorizer()
X = vectorizer.fit_transform(corpus)
def hybrid_chatbot(user_input):
rule_based_responses = {
"hello": "Hello! How can I assist you today?",
"hi": "Hi there! What can I do for you?",
"bye": "Goodbye! Have a great day!"
}
user_input_lower = user_input.lower()
if user_input_lower in rule_based_responses:
return rule_based_responses[user_input_lower]
else:
user_input_vector = vectorizer.transform([user_input])
similarities = cosine_similarity(user_input_vector, X)
response_index = similarities.argmax()
return corpus[response_index]
# Test the chatbot
while True:
user_input = input("You: ")
if user_input.lower() == "exit":
print("ChatBot: Goodbye! Have a great day!")
break
response = hybrid_chatbot(user_input)
print(f"ChatBot: {response}")Exercise 5: Enhancing the Hybrid Chatbot with More Responses
Task: Enhance the hybrid chatbot by adding more rule-based responses and expanding the retrieval-based corpus.
Solution:
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity
corpus = [
"Our office is open from 9 AM to 5 PM, Monday to Friday.",
"We are located at 123 Main Street, Anytown, USA.",
"We offer web development, mobile app development, and digital marketing services.",
"You can contact us at support@company.com.",
"Our phone number is (123) 456-7890."
]
vectorizer = TfidfVectorizer()
X = vectorizer.fit_transform(corpus)
def enhanced_hybrid_chatbot(user_input):
rule_based_responses = {
"hello": "Hello! How can I assist you today?",
"hi": "Hi there! What can I do for you?",
"what is your phone number?": "Our phone number is (123) 456-7890.",
"what is your email address?": "You can contact us at support@company.com.",
"bye": "Goodbye! Have a great day!"
}
user_input_lower = user_input.lower()
if user_input_lower in rule_based_responses:
return rule_based_responses[user_input_lower]
else:
user_input_vector = vectorizer.transform([user_input])
similarities = cosine_similarity(user_input_vector, X)
response_index = similarities.argmax()
return corpus[response_index]
# Test the chatbot
while True:
user_input = input("You: ")
if user_input.lower() == "exit":
print("ChatBot: Goodbye! Have a great day!")
break
response = enhanced_hybrid_chatbot(user_input)
print(f"ChatBot: {response}")These exercises provide hands-on experience with different types of chatbots, reinforcing the concepts covered in this chapter.