Research And Realization Of Automatic Solving Technique For Algebraic Problems In Human-like Answering System

In recent years,many experts and scholars at home and abroad have devoted a lot of energy to researching artificial intelligence,which has promoted the rapid development of artificial intelligence in various fields.Combining artificial intelligence with modern education is also a hot topic for many experts and scholars.The use of artificial intelligence to provide students with personalized,all-round,high-quality educational services and equal educational resources is of great significance to the development of modern education.This paper focuses on the original word processing,based on the computer automatically solve the output of human answer process.Mainly for the following aspects of research and implementation.1.The knowledge representation of algebraic problems and the study of a generic problem solving algorithmKnowledge representation and problem solving algorithm is the premise and basis for problem solving.How to express the algebraic part of the knowledge in a simple and clear way is the premise of the whole problem-solving system.This article adopts an object-oriented knowledge representation method to represent mathematical knowledge in objects.At the same time,the knowledge is divided into two categories: Data and Relation.Data is used to represent the knowledge of atomic types,and Relation is the relationship between one or more Data.It can make the information of knowledge as full as possible.The general problem-solving algorithm is the basis for problem solving.All mathematical problems are solved with their own fixed algorithm.This article first abstracts the solution algorithm through a large number of topics;then these algorithms are implemented in rules with the appropriate granularity.These preparations form the basis of human-like answers.2.Research on Automatic Reasoning of Algebraic ProblemsSolving the problem is the core of the entire system.How to solve a problem completely is the goal of the system.The core of problem solving is knowledge reasoning.There are two main reasoning methods in this paper.One is computational reasoning for multiple-choice and blank-filling questions.The core is the semi-algebraic solution system.The multiple-choice and fill-in questions have features that do not require a solution process.We can use the calculations to quickly calculate the results and avoid the use of time-consuming rule inference.The second method is rule-based reasoning for all types of questions.It is universal.The rules are divided into four categories according to their functions: computational inference rules,logical inference rules,associative inference rules,and rule flows.These four rules perform their duties and jointly complete the task of solving problems with rules.3.Design and Implementation of Automatic Solving System for Algebraic ProblemsThe entire system first expresses the algebraic part of the knowledge entity object,and then through the natural language understanding module,the preprocessing module,the problem-solving module,the shutdown module and the human answering result output module,these five modules cooperate with each other to form a complete algebraic automatic.Solve the system.Finally,through the test,the correctness of the problem for the algebraic part of the system is about 75%,and the average answer time is also within 10 seconds,basically completing the goals set by the initial system design.