Simple 4-layer CNN on raw candlestick charts achieves 0.892 AUC-ROC for cryptocurrency regime prediction and outperforms complex encodings and larger pretrained models.
Deep learning with long short-term memory networks for financial market predictions
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
An integrated node transformer and BERT sentiment model reports 0.80% MAPE for one-day stock predictions on 20 S&P 500 stocks, beating ARIMA and LSTM baselines.
citing papers explorer
-
Visual Chart Representations for Cryptocurrency Regime Prediction: A Systematic Deep Learning Study
Simple 4-layer CNN on raw candlestick charts achieves 0.892 AUC-ROC for cryptocurrency regime prediction and outperforms complex encodings and larger pretrained models.