Effect of training fragment length on Transformers in text complexity prediction.
Rafik HachanaVladimir V. IvanovPublished in: Tiny Papers @ ICLR (2023)
Keyphrases
- prediction accuracy
- text documents
- database
- information retrieval
- multi layer perceptron
- computational complexity
- training set
- space complexity
- explanatory power
- radial basis function network
- training process
- prediction model
- text mining
- worst case
- keywords
- web documents
- training examples
- training samples
- decision problems
- text retrieval
- free text
- prediction error
- computational cost
- training phase
- training data
- prediction algorithm
- total length
- data mining