Detection of effective factors on the adult Health Literacy level with a meta-heuristic algorithm

Health literacy is a global factor in determining the level of social literacy of persons in the individual and social health. To improve health in a community, it is essential which measuring and identifying effective factors on the health literacy. The purpose of this study was to identify and determine the factors affecting the level of health literacy using a meta-heuristic algorithm.  

In this paper, a hybrid model has been introduced which benefit of bat optimization algorithm and machine learning algorithms to determine the major factors on health literacy level. In our model, the support vector machine algorithm and K-nearest neighbor are used to classify the data. The data set has been extracted from a descriptive-analytic study that was performed on 2133 samples to measure adults' health in South Khorasan Province in 2016.

In this study, the combination of bat optimization algorithm and K-nearest neighbor with 93.76% accuracy and 95.47% precision has been able to identify effective factors on health literacy. Combined bat optimization algorithm and support vector machine have been obtained with 96.25 percent accuracy and 97.99 percent precision. Analytical results show that two areas: access to the resources and use of resources are impressive in the low level of health literacy in South Khorasan province are effective.

Considering the performance of the intelligent proposed model show this tool can be used to analyze health literacy data. This model does not require statistical analysis, and can provide high-quality results quickly.