AI Document Search Engine Project

"""Rank a small document collection for a natural-language query."""

from __future__ import annotations

from dataclasses import dataclass
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity


DOCUMENTS = [
    ("python", "Python training covers programming foundations, functions, files, testing, and portfolio projects."),
    ("data", "Data analytics training covers SQL, Excel, dashboards, metrics, Power BI, and business reporting."),
    ("ai", "Artificial intelligence training covers natural language processing, retrieval, model evaluation, automation, and responsible AI."),
    ("cloud", "Cloud training covers compute, storage, identity, networking, monitoring, and deployment practice."),
]


@dataclass(frozen=True)
class SearchResult:
    document_id: str
    text: str
    score: float


class DocumentSearch:
    def __init__(self, documents=DOCUMENTS):
        self.documents = list(documents)
        self.vectorizer = TfidfVectorizer(stop_words="english", ngram_range=(1, 2), sublinear_tf=True)
        self.matrix = self.vectorizer.fit_transform([text for _, text in self.documents])

    def search(self, query: str, limit: int = 3) -> list[SearchResult]:
        if limit < 1:
            return []
        scores = cosine_similarity(self.vectorizer.transform([query]), self.matrix)[0]
        order = scores.argsort()[::-1][:limit]
        return [SearchResult(self.documents[i][0], self.documents[i][1], float(scores[i])) for i in order if scores[i] > 0]


def main() -> None:
    engine = DocumentSearch()
    for result in engine.search("learn SQL and Power BI dashboards"):
        print(result)


if __name__ == "__main__":
    main()