q4_0第一版

最新推荐文章于 2024-09-09 00:00:00 发布

feymanP

最新推荐文章于 2024-09-09 00:00:00 发布

阅读量21

点赞数

文章标签： python 机器学习开发语言

本文链接：https://blog.csdn.net/weixin_43565459/article/details/134890784

版权

guardian_df  = pd.read_csv("Q1_backup.csv")


bad_words = ["plung", "crash", "drop", "tumble", "retreat", "low", "lose", "slide", "fizzle", "sell", "fear"]
good_words = ["buy", "rebound", "climb", "rise", "boom", "buy", "high", "surge", "jump", "soar", "grow"]

guardian_df['positive_title'] = guardian_df['webTitle'].str.contains('|'.join(good_words), case=False, regex=True).astype(int)
guardian_df['negative_title'] = guardian_df['webTitle'].str.contains('|'.join(bad_words), case=False, regex=True).astype(int)

total_positive_titles = guardian_df['positive_title'].sum()
total_negative_titles = guardian_df['negative_title'].sum()

print(f"Total number of positive titles: {total_positive_titles}")
print(f"Total number of negative titles: {total_negative_titles}")

titles_both_positive_and_negative = guardian_df[(guardian_df['positive_title'] == 1) & (guardian_df['negative_title'] == 1)].shape[0]
print(f"Number of titles that are both positive and negative: {titles_both_positive_and_negative}")

aaadsadas