Survival Prediction by an Integrated Learning Criterion on Intermittently Varying Healthcare Data

Authors: Jianfei Zhang, Lifei Chen, Alain Vanasse, Josiane Courteau, Shengrui Wang

AAAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments on the healthcare data demonstrate the effectiveness and generalizability of our model and its promise in practical applications.
Researcher Affiliation Academia 1PROSPCTUS Lab, Department of Computer Science, University of Sherbrooke, Canada 2School of Mathematics and Computer Science, Fujian Normal University, China 3PRIMUS Research Group, Department of Family Medicine, University of Sherbrooke, Canada
Pseudocode No The paper does not contain any structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any concrete access information for source code, such as a repository link or an explicit statement about code release.
Open Datasets No We investigate SPH model on an IV dataset derived from a COPD data collected from Centre Hospitalier Universitaire de Sherbrooke (CHUS).
Dataset Splits Yes In all experiments we report generalized 10CV results, over 100 replicates, in the form of mean standard deviation. The regularizers of this model were estimated via tenfold cross-validation (10CV) in our study. The regularization parameters of SPH, λ1 and λ2, were selected by another 10CV on the training data.
Hardware Specification No The paper does not provide specific hardware details (e.g., CPU/GPU models, memory specifications) used for running its experiments.
Software Dependencies No The paper does not provide specific ancillary software details with version numbers, such as programming language versions or library versions.
Experiment Setup Yes The regularizers of this model were estimated via tenfold cross-validation (10CV) in our study. The Epanechnikov kernel function with a bandwidth value of 1.5 was adopted in the experiment... The kernel bandwidth was set as N 0.2 = 0.3... The regularization parameters of SPH, λ1 and λ2, were selected by another 10CV on the training data. The iterative learning process begins with a warm-start βt1 = (0.5, . . . , 0.5)V and runs until convergence, i.e., until the change of coefficients between two successive iterations is smaller than 10 3.