Research and prediction of opioid crisis based on BP neural network and Markov chain

• DOI: 10.3934/math.2019.5.1357
• Publication Date: 2019-09-05T03:12:08Z
• Authors: Wanchun Fan; ; Yan Jiang; Songyang Huang; Weiguo Liu
• Keywords: grey correlation analysis; 03 medical and health sciences; 0302 clinical medicine; Markov chain; sequential cluster method; QA1-939; 1. No poverty; opioid crisis; BP neural network; 0101 mathematics; 01 natural sciences; Mathematics; 3. Good health

Nowadays, in the United States, opioid abuse is so serious that it has become a crisis, causing health impacts and huge losses to US economy. To study social economic data relationship between drug situation, this paper uses Grey Relation Analysis analyze identification counting of synthetic opioids heroin provided by NFLIS related socioeconomic factors States Census Bureau. After that, orderly clustering used introduce corresponding level. Then, BP neural network Markov model are built forecast degree flood opioid. The Proportion Low Education Level People, Number New Pregnant Women, Elderly Living Alone other 7 were selected as input nodes network. Based on prediction results network, Chain correct residual sequence. It found deviation reduced from[-10.99%, 22.33%] to[-8.29%, 2.81%], making modified value closer measured value. This method combines advantages which improves accuracy provides certain reference values for prediction. Finally, we propose strategies address crisis test their effectiveness.