A novel time series link prediction method: Learning automata approach

The ability to predict linkages among data objects is central to many data mining tasks, such as product recommendation and social network analysis. Substantial literature has been devoted to the link prediction problem either as an implicitly embedded problem in specific applications or as a generic data mining task. This literature has mostly adopted a static graph representation where a snapshot of the network is analyzed to predict hidden or future links. However, this representation is only appropriate to investigate whether a certain link will ever occur and does not apply to many applications for which the prediction of the repeated link occurrences is primary interest (e.g., communication network surveillance). In the time-series link prediction problem, the time series link occurrences are used to predict link occurrence at a particular time. In this paper, we propose a new time series link prediction based on learning automata. In the proposed algorithm for each link that must be predicted there is a learning automaton and each learning automaton tries to predict the existence or not existence of the corresponding link. To predict the link occurrence in time T, there is a chain consists of states 1 through T-1 and the learning automaton passes from state 1 through T-1 to learn the existence or not existence of the corresponding link. Using three coauthorship data sets, we have demonstrated that time-series models of link occurrences achieve better link prediction performance with commonly used static graph link prediction algorithms.