GINN: gradient interpretable neural networks for visualizing financial texts.
Tomoki ItoHiroki SakajiKiyoshi IzumiKota TsubouchiTatsuo YamashitaPublished in: Int. J. Data Sci. Anal. (2020)
Keyphrases
- neural network
- multi layer
- artificial neural networks
- pattern recognition
- neural network model
- edge detection
- fuzzy logic
- gradient information
- genetic algorithm
- natural language
- interactive visualization
- rule extraction
- neural nets
- fuzzy systems
- multilayer perceptron
- stock market
- recurrent neural networks
- training process
- activation function
- financial crisis
- fraud detection
- natural language text
- training algorithm
- gradient direction
- financial information
- artificial neural networks to predict
- associative memory
- radial basis function
- fault diagnosis
- computational intelligence
- association rules
- data analysis
- image segmentation
- decision making
- learning algorithm
- information retrieval
- machine learning