启发式搜索在数学智能解题系统中的应用研究

Application and Research of a Heuristic Search in Intelligent Mathematics Problem Solving System

在人工智能领域,对问题求解的方法都需要某种对解答的搜索,即为一个搜索过程。文中论述的数学智能辅导系统采用了与或树知识表示方法,也可称为问题规约法。它把初始问题通过一系列变换最终变为一个子问题集合,而这些子问题的解可以直接得到,从而解答了初始问题。系统使用以推理深度作估价函数的启发式搜索,使得问题的求解更加有效与合理。论述了采用启发式搜索的必要性及可行性。对比了采用启发式搜索前后,系统解题合理性得到很大提高。

In the field of AI,problem solving corresponds to a search for answers,which is a search process.The intelligent tutoring system discussed by this paper has used a knowledge representation named AND/OR tree,which is also called problem reduction.It decomposes the problem into several subproblems which can be resolved directly,thus the problem can be resolved too.The system has used a heuristic search which uses the ratiocination depth as its heuristic function.Dicussed the necessity and feasibility of the heuristic search.By contrasting the effects,the system obtains more rational solution.