Association Rule Transfer Learning Based on Markov Logic Network

An association rule transfer learning method based on Markov logic networks is presented specific to the situation wherein shared knowledge between the source domain and the target domain is associated with knowledge containing rules, structure, and logic. Having applied this method by means of a pseudo log-likelihood function, the knowledge in the source domain expressed in a Markov logic network is transferred into the target domain while the link between the two domains is established. By means of a self-diagnosis and structure update in the source domain and a new clause surf in the target domain, the mapped structure is optimized so that it can be adapted to learning in the target domain. The experimental results show that the given algorithm successfully maps the transferred knowledge, and improves the precision of the learning model.