Research On Machine Solution Of Probability Problem That Satisfies Hypergeometric Distribution

In recent years,intelligent problem-solving has gradually become a hot topic of artificial intelligence.At present,the study on machine solution for primary mathematics problem has achieved rich results.However,there are not many methods and theories of machine solution for high school mathematics problem.In order to establish a perfect intelligent tutoring system,the research of the machine solution for the probability questions in high school is essential.Among the various probability problems,there is a special kind of probability model called Hypergeometric distribution,which has definite model definition and calculation formula.The objective of this paper is to set up a set of solutions to solve the problem which satisfies Hypergeometric distribution with some established requirements automatically.Most of the high school mathematics problem can be summed up as a calculation to one or a few formulas.The calculation of Hypergeometric distribution problem can be summed up as the probability distribution of a component.Therefore,in this paper,we decompose the process of solving the problem into two parts.The first step is to determine the corresponding formula of the problem.The second step is to extract and supplement the data needed to solve the formula.We turn the solution to the problem into a calculation of the formula.Specifically from the following aspects:First of all,we find the calculation formula of the problem by judging the type of it.A math problem which is in accordance with the model of Hypergeometric distribution can be solved with the calculation formula of Hypergeometric distribution.In this paper,we judge the type of the problem by template matching.We use Regular expression to establish a template library according to the structure information of the part of speech of the words,and get the type of the problem from the template library.Our template library is combined by template primitives automatically,which improves the matching efficiency.At the same time,we establish the auxiliary judgment formula based on the keyword characteristics,which combine with the existence of keywords and its quantitative criteria.Our method makes better use of the information of the problem,and improves the accuracy of the type judgment.Secondly,we establish a method to extract and supplement the relevant data.When the problem solving formula is determined,we need to extract and supplement the data for the solving of this formula.In this paper,we propose a new algorithm to build a complete data list.The execution of the algorithm consists of two nested loops,which are automatically extracted and supplemented data according to the extract rules.In this paper,the problem is divided into "fixed quantity" and "question".We calculate all the possible "question" based on the "fixed quantity" section,and extract the correct answer from the answer set based on the key words in the "question".Experiments show that the proposed machine solution for problems that satisfy Hypergeometric distribution is feasible.