Identifying Stroke Indicators Using Rough Sets

Identifying Stroke Indicators Using Rough Sets

With the spirit of reproducible research, this repository contains all the codes required to produce the results in the manuscript:

Pathan, M. S., Zhang, J., John, D., Nag, A. and Dev, S.(2020). Identifying Stroke Indicators Using Rough Sets, under review.

All codes are written in MATLAB.

Code

• ./Figure3.m: Computes the impact of the dataset size on the correlation value (b/t impact score and accuracy).
• ./Table2_Figure1.m: Computes the performance of the different individual features of electronic health records for detecting stroke.
• ./Table3.m: Computes the (our proposed) impact factor scores for the different individual features of electronic health records.
• ./Table4_Figure2.m: Computes the benchmarking scores and scatter-plots for the different benchmarking approaches.
• ./data/: This folder contains our input data.
• ./results/: This folder will save all the results.
• ./scripts/: This folder contains helper .m files that are necessary for the computation of the different results in the manuscript.

These .m files use the following user-defined helper scripts.

Scripts

• bimodality.m: Computes the bimodality score of a feature vector.
• find_scores.m: Computes the precision, recall, f-score and accuracy values.
• impact_factor.m: Computes the impact factor scores
• impactfactor_from_data.m: Computes the impact factor from the data matrix. The script impact_factor.m is a subset of this file.
• indiscernibility_values_extraction_for_conditional_attributes.m: Computes the indiscernibility values for the conditional attributes.
• indiscernibility_values_extraction_for_decisional_attribute.m: Computes the indiscernibility values for decisional attribute.
• l_factors.m: Computes the loading factor scores for the different features from the input data.