텍스트 요약기

# Extractive text summarization (TF-based)
import re
from collections import Counter

STOP_WORDS = {
    "the", "a", "an", "and", "or", "but", "in", "on", "at", "to", "for",
    "of", "with", "by", "from", "is", "was", "are", "were", "be", "been",
    "has", "have", "had", "do", "does", "did", "will", "would", "could",
    "should", "that", "this", "it", "its", "he", "she", "they", "we", "you",
    "i", "not", "no", "as", "if", "so", "than", "then", "more", "most",
}

def tokenize_sentences(text: str) -> list[str]:
    sentences = re.split(r'(?<=[.!?])\s+', text.strip())
    return [s.strip() for s in sentences if len(s.strip()) > 10]

def word_frequency(sentences: list[str]) -> dict[str, int]:
    words = re.findall(r"[a-z']+", " ".join(sentences).lower())
    return Counter(w for w in words if w not in STOP_WORDS and len(w) > 2)

def score_sentences(sentences: list[str], freq: dict[str, int]) -> list[float]:
    scores = []
    n = len(sentences)
    for i, sentence in enumerate(sentences):
        words = re.findall(r"[a-z']+", sentence.lower())
        score = sum(freq.get(w, 0) for w in words)
        if words:
            score /= len(words)  # normalize by length
        # Position weight
        rel_pos = i / max(n - 1, 1)
        if rel_pos <= 0.2:
            score *= 1.4
        elif rel_pos >= 0.8:
            score *= 1.2
        scores.append(score)
    return scores

def summarize(text: str, num_sentences: int = 3) -> str:
    sentences = tokenize_sentences(text)
    if len(sentences) <= num_sentences:
        return text

    freq = word_frequency(sentences)
    scores = score_sentences(sentences, freq)

    # Get top-N sentence indices, sort by original position
    ranked = sorted(range(len(scores)), key=lambda i: -scores[i])[:num_sentences]
    selected = sorted(ranked)

    return " ".join(sentences[i] for i in selected)

# Example
text = """..."""  # Your long text here
print(summarize(text, num_sentences=3))