AI Intent Classification Project

"""Classify short assistant requests into supported intents."""

from __future__ import annotations

from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.linear_model import LogisticRegression
from sklearn.pipeline import Pipeline


TRAINING_DATA = {
    "greeting": ["hello", "hi there", "good morning", "hey assistant", "good evening", "nice to meet you"],
    "course_info": ["tell me about the course", "what topics are covered", "show the syllabus", "course duration please", "what will I learn", "share course details"],
    "fees": ["what is the fee", "how much does training cost", "course price", "payment details", "tell me the tuition", "is there an installment option"],
    "schedule": ["when is the next batch", "class timings", "weekend schedule", "what time are classes", "batch start date", "weekday training time"],
}


def train_classifier() -> Pipeline:
    texts, labels = [], []
    for label, examples in TRAINING_DATA.items():
        texts.extend(examples)
        labels.extend([label] * len(examples))
    model = Pipeline([
        ("tfidf", TfidfVectorizer(ngram_range=(1, 2), lowercase=True)),
        ("classifier", LogisticRegression(max_iter=1000, random_state=42)),
    ])
    return model.fit(texts, labels)


def classify(model: Pipeline, text: str) -> tuple[str, float]:
    probabilities = model.predict_proba([text])[0]
    index = int(probabilities.argmax())
    return str(model.classes_[index]), float(probabilities[index])


def main() -> None:
    model = train_classifier()
    for message in ("Hello assistant", "How much is the course?", "When does the weekend batch start?"):
        label, confidence = classify(model, message)
        print(f"{message!r}: {label} ({confidence:.3f})")


if __name__ == "__main__":
    main()