• Barabasi-Albert Network
• Classification
• Epidemic model
• Erdos-Renyi Network
• Quantification
• Quantifying Diffusion Process
• Stochastic Network
• The SI Model
• The SIR Model
• The SIS Model
• Watts-Strogatz Network

The diffusion and spreading of ideas, innovation, information, influence, viruses and diseases are ubiquitous in social and information networks. Accordingly, This study focuses on the diffusion process of epidemic model classification and quantification in stochastic network aims to accurately predict the state transition of a node in the network given its present state. Thus, Epidemic models selected for diffusion process in the network and these epidemic models diffused on stochastic networks. As a result, the data classification applied on a decision tree classifier and the evolution of the network quantified by Kullback-Leibler Divergence. In the SI epidemic model, all the three types of random network obtained similar pattern of f1-score. Lower infection rate yields a high f1-score. According to KL divergence, the quantification of the actual and estimated distribution is bounded by zero and one. In the SIS epidemic model, the f1-score depend on the probability of state transition of infection and recovery rate. when the recovery rate is lower than infection rate while predicting the next states of the nodes in all the three stochastic networks leads to f1-score close to one. Moreover, the diffusion of SIS model on Barabasi- Albert, Erdos-Renyi and Watts-Strogatz random network quantified through KL divergence and its outcome is in between zero and one. In the SIR epidemic model, f1-score depend on the state transition probability of infection and removal values. while infection rate is greater than removal rate higher value of f1-score obtained regardless of the stochastic network. The actual and estimated distribution quantified through KL divergence obtained a result between zero and one irrespective of the random networks.