nerExtraction.py

import sys
import io
import re
from kiwipiepy import Kiwi
from flask import Flask, request, jsonify

#출력 인코딩을 UTF-8로 설정
sys.stdout = io.TextIOWrapper(sys.stdout.buffer, encoding='utf-8')

from gliner import GLiNER

app = Flask(__name__)
model = GLiNER.from_pretrained('urchade/gliner_multi')


@app.route('/nerExtraction', methods=['POST'])
def ner_extraction():
    data = request.json
    input_text = data['text']
    labels = data['labels']

    entities = model.predict_entities(input_text, labels)
    entities_ref = {entity["label"]:[noun_extractor(entity["text"])] for entity in entities}

    #정규 표현식으로 '주X일' 또는 'X일' 패턴 찾기
    pattern = r"(주?\d+일)"

    work_periods = re.findall(pattern, input_text)

    #결과에 추가
    entities_ref['WorkPeriod'] = work_periods
    return jsonify(entities_ref)


def noun_extractor(text):
    results = []
    kiwi = Kiwi()
    result = kiwi.analyze(text)
    for token, pos, _, _ in result[0][0]:
        if len(token) != 1 and (pos.startswith('N') or pos.startswith('SL')):
            results.append(token)
    return results

if __name__ == '__main__':
    app.run(host='0.0.0.0', port=5000, debug=True)