Taxonomy is an important component in knowledge bases, and it is an urgent, meaningful but challenging task for Chinese taxonomy construction. In this paper, we propose a taxonomy induction approach from a Chinese encyclopedia by using combinatorial optimizations. At first,
relations are derived by validating the relation between two categories. Then, integer programming optimizations are applied to find out
relations from encyclopedia articles by considering the constrains among categories. The experimental results show that our approach can construct a practicable taxonomy from Chinese encyclopedias.