Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Do Weibo platform experts perform better at predicting stock market?

Ziyuan Ma, Conor Ryan, Jim Buckley, Muslim Chochlov

TL;DR

This paper investigates how a user’s financial background influences sentiment-based stock market prediction on the Weibo platform. It combines a BERT-based sentiment classifier with a two-layer LSTM stock predictor, applied separately to sentiments from Authorized Financial Advisors (AFA) and Unauthorized Financial Advisors (UFA), using Hang Seng Index data from 2018–2019. The study finds that AFA-derived predictions are markedly more accurate, achieving about a 39.67% precision improvement and up to 87% accuracy compared with existing approaches, while UFA signals underperform. These results suggest that expert financial content yields more predictive sentiment signals, with potential implications for weighting or filtering social-media sentiment in forecasting models and for future analyses distinguishing different authority levels.

Abstract

Sentiment analysis can be used for stock market prediction. However, existing research has not studied the impact of a user's financial background on sentiment-based forecasting of the stock market using artificial neural networks. In this work, a novel combination of neural networks is used for the assessment of sentiment-based stock market prediction, based on the financial background of the population that generated the sentiment. The state-of-the-art language processing model Bidirectional Encoder Representations from Transformers (BERT) is used to classify the sentiment and a Long-Short Term Memory (LSTM) model is used for time-series based stock market prediction. For evaluation, the Weibo social networking platform is used as a sentiment data collection source. Weibo users (and their comments respectively) are divided into Authorized Financial Advisor (AFA) and Unauthorized Financial Advisor (UFA) groups according to their background information, as collected by Weibo. The Hong Kong Hang Seng index is used to extract historical stock market change data. The results indicate that stock market prediction learned from the AFA group users is 39.67% more precise than that learned from the UFA group users and shows the highest accuracy (87%) when compared to existing approaches.

Do Weibo platform experts perform better at predicting stock market?

TL;DR

This paper investigates how a user’s financial background influences sentiment-based stock market prediction on the Weibo platform. It combines a BERT-based sentiment classifier with a two-layer LSTM stock predictor, applied separately to sentiments from Authorized Financial Advisors (AFA) and Unauthorized Financial Advisors (UFA), using Hang Seng Index data from 2018–2019. The study finds that AFA-derived predictions are markedly more accurate, achieving about a 39.67% precision improvement and up to 87% accuracy compared with existing approaches, while UFA signals underperform. These results suggest that expert financial content yields more predictive sentiment signals, with potential implications for weighting or filtering social-media sentiment in forecasting models and for future analyses distinguishing different authority levels.

Abstract

Sentiment analysis can be used for stock market prediction. However, existing research has not studied the impact of a user's financial background on sentiment-based forecasting of the stock market using artificial neural networks. In this work, a novel combination of neural networks is used for the assessment of sentiment-based stock market prediction, based on the financial background of the population that generated the sentiment. The state-of-the-art language processing model Bidirectional Encoder Representations from Transformers (BERT) is used to classify the sentiment and a Long-Short Term Memory (LSTM) model is used for time-series based stock market prediction. For evaluation, the Weibo social networking platform is used as a sentiment data collection source. Weibo users (and their comments respectively) are divided into Authorized Financial Advisor (AFA) and Unauthorized Financial Advisor (UFA) groups according to their background information, as collected by Weibo. The Hong Kong Hang Seng index is used to extract historical stock market change data. The results indicate that stock market prediction learned from the AFA group users is 39.67% more precise than that learned from the UFA group users and shows the highest accuracy (87%) when compared to existing approaches.
Paper Structure (18 sections, 1 equation, 3 figures, 2 tables)

This paper contains 18 sections, 1 equation, 3 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related work
  3. Sentiment analysis using neural networks
  4. Machine learning in stock market prediction
  5. Stock market prediction system
  6. Data collection module
  7. Data cleaning module
  8. Sentiment analysis module
  9. LSTM stock prediction module
  10. Implementation details
  11. Methodology
  12. User grouping method
  13. Daily sentiment analysis method
  14. Time window calculation method
  15. Stock prediction model evaluation metrics
  16. ...and 3 more sections

Figures (3)

  • Figure 1: Stock market prediction system structure
  • Figure 2: Pearson's correlation for different T values.
  • Figure 3: Predict price