Mynli

//pip install nltk scikit-learn numpy pandas

import nltk
from nltk.stem import WordNetLemmatizer
import numpy as np
import random
import string  # to process standard python strings

nltk.download('punkt')
nltk.download('wordnet')

lemmatizer = WordNetLemmatizer()

# Example preprocessing function
def preprocess_input(text):
    text = text.lower()  # Convert to lowercase
    tokens = nltk.word_tokenize(text)  # Tokenize the input
    tokens = [lemmatizer.lemmatize(token) for token in tokens if token not in string.punctuation]  # Lemmatize and remove punctuation
    return tokens

training_data = [
    {"intent": "budgeting", "patterns": ["How do I create a budget?", "What is a budget?", "Budgeting tips"]},
    {"intent": "investing", "patterns": ["How to invest money?", "Best investments for beginners", "How does investing work?"]},
    {"intent": "saving", "patterns": ["How do I save more money?", "Best saving strategies", "How much should I save?"]},
    {"intent": "credit", "patterns": ["How do credit scores work?", "How can I improve my credit score?", "What affects my credit score?"]},
]

categories = ["budgeting", "investing", "saving", "credit"]

from sklearn.feature_extraction.text import CountVectorizer
from sklearn.naive_bayes import MultinomialNB

# Prepare data for training
def prepare_training_data(data):
    texts = []
    labels = []
    for item in data:
        for pattern in item["patterns"]:
            texts.append(pattern)
            labels.append(item["intent"])
    return texts, labels

texts, labels = prepare_training_data(training_data)

# Convert text data into numerical features
vectorizer = CountVectorizer(tokenizer=preprocess_input)
X = vectorizer.fit_transform(texts)
y = np.array(labels)

# Train the Naive Bayes model
model = MultinomialNB()
model.fit(X, y)

responses = {
    "budgeting": "To create a budget, start by listing your income and expenses. Then, allocate your money based on your priorities and track your spending regularly.",
    "investing": "Investing involves putting money into assets like stocks, bonds, or real estate in order to grow your wealth over time.",
    "saving": "A good saving strategy is to set aside a percentage of your income each month. Aim for at least 20% if possible, but even small amounts can build up over time.",
    "credit": "Credit scores are calculated based on your payment history, the amount of credit you use, and how long you’ve had credit accounts."
}

def chatbot_response(user_input):
    user_input = [user_input]
    X_user = vectorizer.transform(user_input)  # Convert user input into vector form
    predicted_intent = model.predict(X_user)[0]  # Predict the intent category

    # Fetch the appropriate response
    return responses.get(predicted_intent, "I'm not sure how to answer that. Can you ask another financial question?")

print("Mynli: Hi! I'm Mynli, your financial literacy assistant. How can I help you today?")
while True:
    user_input = input("You: ")
    if user_input.lower() == "exit":
        print("Mynli: Goodbye!")
        break
    response = chatbot_response(user_input)
    print(f"Mynli: {response}")