Stock Chart Pattern recognition with Deep Learning
arXiv:1808.00418v130 citations
Originality Synthesis-oriented
AI Analysis
This work addresses pattern recognition for stock traders, but it is incremental as it applies existing deep learning methods to financial data.
The study tackled stock chart pattern recognition by evaluating CNN and LSTM performances on historical data, reporting specific accuracies for identifying common patterns.
This study evaluates the performances of CNN and LSTM for recognizing common charts patterns in a stock historical data. It presents two common patterns, the method used to build the training set, the neural networks architectures and the accuracies obtained.