The NLP Sandbox: an efficient model-to-data system to enable federated and unbiased evaluation of clinical NLP models.
Yao YanThomas YuKathleen MuenzenSijia LiuConnor BoyleGeorge KoslowskiJiaxin ZhengNicholas J. DobbinsClement EssienHongfang LiuLarsson OmbergMeliha YestigenBradley TaylorJames A. EddyJustin GuinneySean D. MooneyThomas SchaffterPublished in: CoRR (2022)
Keyphrases
- experimental data
- probabilistic model
- prior knowledge
- predictive model
- learned models
- input data
- accurate models
- learning models
- statistical methods
- natural language processing
- models built
- data analysis
- measured data
- subject specific
- classification models
- autoregressive
- test data
- raw data
- evaluation method
- historical data
- data sets
- temporal dependencies
- data sources
- database
- statistical analysis
- training data
- discrete data
- data points
- modeling framework
- parametric models
- hidden variables
- relational models
- neural network model
- statistical models
- missing data
- dimensional vector
- parameter estimation
- computational models
- hybrid model
- statistical model
- autoregressive moving average
- natural language
- computational model
- knowledge discovery
- digital libraries
- clinical practice
- knowledge representation
- linear model
- information extraction
- text mining
- question answering
- mathematical model