Strategies Prediction and Combination of Multi-strategy Ontology Mapping

Most of current ontology mapping methods can not treat different mapping tasks in different ways referred to the features of the input ontology. And they combine different mapping strategies without full consideration of the influences on mapping results caused by mapping strategies. In view of the above questions, this paper proposes a dynamic mapping policy selection method which can pre-select mapping strategy and use entropy decision-making method to determine the combined weight of the selected strategy. Experiments show that this method can maintain the stability and the commonality, and improve the recall ratio and the precision ratio at the same time.