Table 5.

Interrelation between AI and market efficiency.

Application purpose	Method	Performance criteria	References
Analysis of SEC reports and investor attention	SEC's EDGAR	The attention of sophisticated investors for the earning announcement impacting on portfolio performance is measured	Li R. et al., 2019
Analysis of endogenous information acquisition	SEC's EDGAR	A long-short portfolio based on different measures of information acquisition activity generates a monthly abnormal return of 80 basis points that is not reversed in the long-run	Li and Sun, 2022
Arbitrage trading strategy based on machine learning	LR, RF, Gradient Boosting Classifier	Volume-Weighted Average Prices (VWAP), ML models outperform the general market by far, which poses a clear challenge to the semi-strong form of market efficiency in futures markets	Waldow et al., 2021
ML algorithms to find profitable technical trading rules using past prices		Genetic algorithm, KNN, RF The out-of-sample profitability decreases through time, becoming the markets more efficient over time	Brogaard and Zareei, 2021
Analysis of cryptocurrency market efficiency	RNN applied to XBTEUR time series bitcoin market	Applying F-measures authors show that Bitcoin market is partially efficient	Hirano et al., 2018
Testing the weak-form efficient market	SVM and LR	Randomness of a sequence of rising/falling states of stock prices	Khoa and Huynh, 2021