With the rapid development of the internet, the classical internet public opinion (IPO) problem as an important social issue has been studied for several years. However, few of models are out of the dimension of differential equations. In this paper, a novel mathematical model for IPO problem based on uncertainty theory is first proposed. A hybrid intelligent algorithm consisted by 99method and an improved genetic algorithm is given to solve the proposed model. Finally, a numerical example

more »

... on a real event is given to show the efficiency and usefulness of the proposed methodology. The result shows that the IPO model achieves good modeling and control performance. Introduction Nowadays, the internet public opinion (IPO) problem becomes a hot issue studied by many scholars. According to China Internet Network Information Center (CNNIC) 39th statistical report on internet development in Beijing (released on the 22nd of January, 2017) it follows that-as of the end of December, 2016, Chinese netizens reached 731 million, the internet penetration rate reached 53.2%, thus about half of the Chinese population have access to the internet. Online social networks create a platform for people to communicate and connect, meanwhile, it has some characteristics by the development of the internet such as openness, complexity, and non integrity. Under the relatively weak management of the network environment, the harmfulness of internet public opinion is magnified out of control. In order to solve this problem, Sudbury [1] first built SIR model based on the biological theory. After that, many scholars have investigated under different complex networks [2] [3] [4] [5] [6] [7] [8] [9] . Meanwhile, Wang [10] applied game theory to the IPO problem model, Liu et al. [11] investigated the IPO problem by cellular automata. Many of the above methods are using probability theory which are based on the existing data to deal with the uncertain factors, the IPO problem, however, exhibits sudden outbreaks with no historical data. In order to deal with such situations, uncertainty theory was introduced by Liu [12] and refined by Liu [13] . Uncertain differential equations [14,15], uncertain processes [16], uncertain programming [17,18], uncertain set theory [19, 20] , among others [21] [22] [23] [24] , have also been proposed. Later, Su [25] described the IPO problem under uncertain differential equation and combined it with real data examples to illustrate the effectiveness of the model. This paper will propose a new mathematical model based on the previous studies and on uncertainty theory. Our model will extend the differential equations model mentioned above. A hybrid intelligent algorithm is proposed here to solve the model, and a real data example to prove its effectiveness. The rest of the paper is organized as follows. Section 2 introduces some basic concepts and theorems from uncertainty theory. In Section 3, the uncertain mathematical model for IPO problem is proposed and explained in detail. In Section 4, the techniques needed for the model are discussed. In Section 5, a numerical example and a real example are given to illustrate the effectiveness of our proposed model. A brief summary is given in Section 6.