Paper: arXiv 2410.07143

Authors: Saber Talazadeh, Dragan Perakovic

Abstract

Stock trend forecasting, a challenging problem in the financial domain, involves ex-tensive data and related indicators. Relying solely on empirical analysis often yields unsustainable and ineffective results. Machine learning researchers have demonstrated that the application of random forest algorithm can enhance predictions in this context, playing a crucial auxiliary role in forecasting stock trends. This study introduces a new approach to stock market prediction by integrating sentiment analysis using FinGPT generative AI model with the traditional Random Forest model. The proposed technique aims to optimize the accuracy of stock price forecasts by leveraging the nuanced understanding of financial sentiments provided by FinGPT. We present a new methodology called “Sentiment-Augmented Random Forest” (SARF), which in-corporates sentiment features into the Random Forest framework. Our experiments demonstrate that SARF outperforms conventional Random Forest and LSTM models with an average accuracy improvement of 9.23% and lower prediction errors in pre-dicting stock market movements.

Complexity vs Empirical Score

  • Math Complexity: 2.5/10
  • Empirical Rigor: 6.5/10
  • Quadrant: Street Traders — empirical and applied, lower math complexity

Why this score: The paper uses established machine learning methods (Random Forest, FinGPT) without novel mathematical derivations, keeping math complexity low, but includes detailed empirical work with specific datasets, APIs, and comparative metrics.

Research Flowchart

  flowchart TD
  Start(["Research Goal: Enhance Stock Trend Prediction"]) --> Inputs
  subgraph Inputs ["Data & Models"]
      direction LR
      I1["(Historical Market Data)"]
      I2["(Financial News via FinGPT)"]
  end
  Inputs --> Method
  subgraph Method ["Key Methodology: SARF"]
      direction TB
      M1["Sentiment Feature Extraction"] --> M2["Feature Fusion<br>Sentiment + Technical Data"]
      M2 --> M3["Random Forest Model Training"]
  end
  Method --> Outcome
  subgraph Outcome ["Findings & Outcomes"]
      direction LR
      O1["9.23% Accuracy Improvement"]
      O2["Lower Prediction Errors"]
      O3["Superior to LSTM/RF"]
  end

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes a Sentiment‑Augmented Random Forest (SARF) model that integrates FinGPT‑derived sentiment features with traditional Random Forest technical indicators for stock prediction.

Source sentence
We present a new methodology called "Sentiment-Augmented Random Forest" (SARF), which incorporates sentiment features into the Random Forest framework.

Assumes 2

  • A fixed random number seed is used to mitigate sampling error. model
    Source sentence
    We also considered a fixed random number seed to mitigate sampling error.
  • FinGPT provides sentiment scores ranging from -1 (negative) to 1 (positive). data
    Source sentence
    we used FinGPT APIs to analyze text data and get a sentiment score ranging from -1 (negative) to 1 (positive).

Data

  • Three stock indices (S&P 500, Nasdaq, Dow Jones) market-prices unknown
    S&P 500, Nasdaq, Dow Jones indices
    Source sentence
    To establish an effective technical data system, we carefully select three indices as the dataset
  • Technical indicators (15) calculated with TA‑Lib · TA‑Lib (Technical Analysis Library) alternative public
    Source sentence
    We used TA-Lib (Technical Analysis Library) to calculate 15 technical indicators.
  • Sentiment indicators (4) from FinGPT · FinGPT text-news public
    Source sentence
    From a combination of 15 technical indicators and 4 sentiment indicators, we derive 14 predictor variables for forecasting.

Recognised sources: Alpha Vantage

Costs

Not applicable The paper reports no trading or execution result.

Finds 3

  • SARF outperforms conventional Random Forest and LSTM models with an average accuracy improvement of 9.23%. performance
    Source sentence
    Our experiments demonstrate that SARF outperforms conventional Random Forest and LSTM models with an average accuracy improvement of 9.23% and lower prediction errors in predicting stock market movements.
  • On the three indices, Optimized Random Forest (SARF) achieves higher accuracy (S&P 500 0.78, Nasdaq 0.85, Dow Jones 0.82) than traditional Random Forest. predictive-accuracy
    Source sentence
    Table 3. Experiments Outcome evaluating the accuracy of models on stock indices
  • Preliminary results demonstrate that SARF outperforms conventional Random Forest models, especially in volatile market conditions. performance
    Source sentence
    Preliminary results demonstrate that SARF outperforms conventional Random Forest models, showcasing its efficacy in predicting stock market movements, particularly in volatile market conditions.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 94 related papers). Coverage is partial, so this is neither replication nor refutation.

Cited by 1 paper in the archive (1 by independent authors)

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.

Cited References (23)

  1. FinGPT: Democratizing Internet-scale Data for Financial Large Language Models (2023) — link
  2. Stock Price Prediction using Sentiment Analysis and Deep Learning for Indian Markets (2022) — link
  3. An Economic Forecasting Method Based on the LightGBM-Optimized LSTM and Time-Series Model (2021) — link
  4. The potential of new ensemble machine learning models for effluent quality parameters prediction and related uncertainty (2020) — link
  5. Financial Trading Strategy System Based on Machine Learning (2020) — link
  6. Stock market prediction using machine learning classifiers and social media, news (2020) — link
  7. A Robust Predictive Model for Stock Price Prediction Using Deep Learning and Natural Language Processing (2019) — link
  8. Cascading logistic regression onto gradient boosted decision trees for forecasting and trading stock indices (2019) — link
and 15 more
  1. Stock selection with random forest: An exploitation of excess return in the Chinese stock market (2019) — link
  2. Random Forest Based Feature Selection of Macroeconomic Variables for Stock Market Prediction (2019) — link
  3. Investor attention and short-term return reversals (2019) — link
  4. Artificial Counselor System for Stock Investment (2019) — link
  5. Global stock market investment strategies based on financial network indicators using machine learning techniques (2019) — link
  6. Classification of intraday S&P500 returns with a Random Forest (2019) — link
  7. Predicting the direction of stock market prices using tree-based classifiers (2019) — link
  8. Indian Stock Market Prediction Using Machine Learning and Sentiment Analysis (2018) — link
  9. Combining the wisdom of crowds and technical analysis for financial market prediction using deep random subspace ensembles (2018) — link
  10. Twitter as a tool for forecasting stock market movements: A short-window event study (2018) — link
  11. Comparing Technical and Fundamental Indicators in Stock Price Forecasting (2018) — link
  12. News-based forecasts of macroeconomic indicators: A semantic path model for interpretable predictions (2018) — link
  13. Impact of financial ratios and technical analysis on stock price prediction using random forests (2017) — link
  14. Combining of random forest estimates using LSboost for stock market index prediction (2017) — link
  15. Evaluating the impact of technical indicators on stock forecasting (2016) — link

0 of 23 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.