Part-of-Speech Tagging with Jieba in Python

Part-of-speech tagging assigns a grammatical category to each word, which is useful for tasks such as keyword extraction, filtering, and analyzing word distribution in texts.

Jieba's POS tagging follows the ICTCLAS-compatible tag set. Below is a simple example that loads stop words, fetches an article using Newspaper3k, performs precise‑mode segmentation with jieba.posseg, filters out stop words, and stores the word‑tag pairs in a pandas DataFrame.

import newspaper
import pandas as pd
import jieba.posseg as pseg

# Load stop words
stopWords = [line.strip() for line in open('stopWord2.txt', encoding='gbk').readlines()]

# Get article (example)
article = newspaper.Article('https://finance.sina.com.cn/money/bank/bank_hydt/2019-02-25/doc-ihsxncvf7656807.shtml', language='zh')
article.download()
article.parse()
article.nlp()
article_words = "".join(article.keywords)

seg_list_exact = pseg.cut(article_words)  # precise mode
words_list = []  # store (word, tag)

for word in seg_list_exact:
    if word not in stopWords:
        words_list.append((word.word, word.flag))

words_pd = pd.DataFrame(words_list, columns=['word', 'type'])
print(words_pd.head())  # display