Збережено в:
| Автори: | , , , , |
|---|---|
| Формат: | Recurso digital |
| Мова: | Португальська |
| Опубліковано: |
Zenodo
2025
|
| Предмети: | |
| Онлайн доступ: | https://doi.org/10.5281/zenodo.15092801 |
| Теги: |
Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
Зміст:
- <p><span>This study investigates the relationship between sentiment expressed in financial news and market movements, using news articles extracted from Yahoo Finance between September 2020 and September 2023. The methodology applied Natural Language Processing (NLP) techniques for sentiment analysis, employing the NLTK (VADER) and TextBlob libraries. News articles were collected through web scraping and analyzed for polarity (positive, neutral, or negative), then compared with fluctuations in the S&P 500 index. The results indicated that financial news sentiment was, on average, slightly positive, with minor discrepancies between analytical methods. In certain periods, such as early 2021, an increase in positive news coincided with market recovery, but the relationship between sentiment and market performance was not linear, suggesting the influence of macroeconomic and geopolitical factors. It is concluded that sentiment analysis can serve as a complementary indicator for predicting financial market trends, but its application should consider multiple sources of information and more advanced models to improve forecast accuracy.</span></p>