Abstract
Security measures have become an important part of operational compliance in healthcare domain. Owing to increasing cyber threats on internet-based healthcare devices, it is essential to provide solution for proactive security planning. This includes understanding security requirements based on historical decisions and finding the right combination for security implementation. For the purpose, we propose a Classification-Based Security Modification (CBSM) framework which helps in predicting the need for change in security setting based on prior history. The framework also compares multiple classifiers like Decision Tree, Naïve Bayes and RandomForest and reports the best classifier. The predictors include: (i) change in technology infrastructure, (ii) change in hospital performance (iii) hospital characteristics and (iv) socio-economic factors. Our work contributes to future research on security investment policy based on hospital performance and motivates managers to proactively improve their security settings based on new technology installations and change in hospital performance.
Recommended Citation
Pal, Shounak and Mukhopadhyay, Arunabha, "A CBSM Framework for requirement-based Proactive Security Measure in Hospitals" (2019). AMCIS 2019 Proceedings. 66.
https://aisel.aisnet.org/amcis2019/treo/treos/66
A CBSM Framework for requirement-based Proactive Security Measure in Hospitals
Security measures have become an important part of operational compliance in healthcare domain. Owing to increasing cyber threats on internet-based healthcare devices, it is essential to provide solution for proactive security planning. This includes understanding security requirements based on historical decisions and finding the right combination for security implementation. For the purpose, we propose a Classification-Based Security Modification (CBSM) framework which helps in predicting the need for change in security setting based on prior history. The framework also compares multiple classifiers like Decision Tree, Naïve Bayes and RandomForest and reports the best classifier. The predictors include: (i) change in technology infrastructure, (ii) change in hospital performance (iii) hospital characteristics and (iv) socio-economic factors. Our work contributes to future research on security investment policy based on hospital performance and motivates managers to proactively improve their security settings based on new technology installations and change in hospital performance.