Research On Evaluation Model Of Junior High School Students' Innovation Ability In Maker Environment

The maker movement,which has swept the world,not only has an impact on the economic and cultural fields,but also set off a wave of epochal significance in the field of education.Gen education as a guest sport and crashed the product of modern education,also in promoting our country's basic education reform in recent years,many primary and secondary schools have set up a guest space,build a guest environment,carry out the education in the teaching and research,along with a guest environment bring us the diversity of education at the same time,more should pay attention to the guest the essence of education.On the one hand,we still have a poor understanding of maker education,ignoring that the core goal of maker education is to cultivate students' maker literacy and improve students' innovation ability.Many schools are limited by resources and conditions,and the construction of maker environment is not perfect enough.On the other hand innovation ability evaluation index system about the guest environment research is not perfect,the existing evaluation methods of too much attention to student's entity works and ignores the whole design process,the evaluation index of subjective factors,the evaluation dimensions is not joint innovation under the age of the purpose of the education in primary and secondary schools to carry out a guest in our country,has a certain bias and limitations.This paper analyzes the essence and connotation of maker education,clarifies the evaluation goal of innovation ability in maker environment,selects evaluation indexes according to the physical and mental development characteristics and cognitive level of junior high school students,calculates the index weight by means of combination weighting,and makes the setting of index weight more in line with the actual situation.With the continuous development of artificial intelligence,considering the characteristic of nonlinear evaluation indexes,the BP neural network on the problem of nonlinear processing indexes,showing good properties,its self-learning adaptive ability can be very good deal with dynamic evaluation index,but its there is easy to fall into local optimal solution and the defects such as slow convergence speed.On the BP neural network easy to fall into local optimal solution,using particle swarm algorithm to optimize design,in the design process of particle swarm optimization(pso)algorithm in the optimization process defects and the insufficiency,the introduction of inertia weight and adaptive mutation factor of particle swarm optimization(pso)algorithm was improved,avoid the scope of the search space is limited to search for the optimal solution,It is helpful to search the optimal weights and thresholds,and accelerate the speed of searching and the accuracy of the results.For innovation ability evaluation,there are more types and various indicators of the characteristics of nonlinear characteristics and the guest of the education teaching mode,target and evaluation method are constantly changing,the complexity of the indicators,the difficulty of the evaluation are prompted the evaluation technology constantly updated,the introduction of improved particle swarm algorithm to optimize the BP neural network,Compared with other network models,the network structure and calculation process are relatively simple.Only the objective function can be considered without other auxiliary operations.In addition,implicit laws can be found by continuous training and learning from complex sample data,so as to avoid falling into the local optimal solution or failing to search for the optimal solution.The evaluation results obtained in the end have the advantages of objectivity and efficiency,strong adaptability and high accuracy,which are superior to the traditional evaluation methods.Finally,three different test functions are used to analyze and test the improved particle swarm optimization algorithm,and the effectiveness and rationality of the improved particle swarm optimization algorithm are proved by the Matlab simulation test results.After that,the collected evaluation data were brought into the three network models for simulation training.By comparing the iteration times,moderation value and accuracy of the experimental results,the advantages of the improved particle swarm optimization algorithm to optimize the BP neural network model were demonstrated.The error results obtained by training in the IPSO-BP model were also relatively ideal.The experimental results can prove that the model can be put into the future evaluation to promote teaching,so as to promote the sustainable development of maker education in junior high schools in China.