Abstract: Quantifying similarity/difference between two objects plays an important role in many contexts The quality of the similarity/difference scores can be improved by considering the semantic information related to the features of objects A flexible semantic distance function called X Dist is proposed, which can utilize the semantic information to measure the difference between two objects based on a solution to the transportation problem from linear optimization With a ground distance function for single features being a metric, X Dist is also a metric This property is very useful for making searching efficient, but is not investigated in the previous research Moreover, the experimental results show X Dist can be as good as the previously studied similarity measures in nearest neighbor searching, discriminative power and computing speed