Application Of Several Data Mining Algorithms In The Analysis Of Ancient Ceramics

Chinese ancient ceramics has become the jewel of the development of human civilization with its exquisite craftsmanship and the social value conveyed.In recent years,with the discovery of a large number of ancient ceramics,the rapid development of ceramic industry and the strong demand for the identification of ancient ceramics,the study of ancient ceramics source dating has become a current research hotspot.At present,the most commonly used methods for the dating analysis of ancient ceramics are the traditional empirical methods and the scientific and technical methods.Traditional experience identification methods rely mainly on human visual touch to identify cultural relics.There are limitations of human senses and experience,and they are susceptible to subjective factors.Scientific and technical identification methods mainly rely on scientific and technological means to identify the age,composition and origin of ancient ceramics,etc.Scientific and technical identification methods mainly include element composition analysis method,thermoluminescence analysis method,X-ray fluorescence spectrometer analysis method,etc,but this method also has some disadvantages,such as damage to the integrity of ancient ceramics,the identification needs a lot of manpower and financial resources.The two methods have their own advantages and disadvantages,and the data mining technology can be used to analyze and study the components of ancient ceramics and find the internal relations and rules,which can make up for the shortcomings of the two methods.With the rapid development of data mining technology,more and more scholars have connected data mining technology with various fields,and formed a very perfect theoretical research system.Based on these theoretical systems,the author,on the basis of in-depth study and study of data mining books and the latest research progress in related fields,through the collection of ancient ceramic chemical composition data analysis,BP neural network,particle swarm optimization(pso),genetic optimization(ga),random forest and LSTM are applied to the study of ancient ceramicsIn the third chapter of this article,the model of the ceramic raw materials is established by BP neural network algorithm.and then the BP neural network was optimized by using genetic algorithm.The experimental results showed that the classification accuracy and accuracy rate of GA-BP was significantly higher than that of BP neural network model.The fourth chapter is based on the ancient ceramic particle swarm algorithm to optimize the BP neural network classification method,the fifth chapter uses the random forest algorithm to classify ancient ceramicss,The LSTM network in deep learning is adopted in chapter 6.The experimental results show that theLSTM network model is the best model for the dating of ancient ceramic source.However,the experimental results do not achieve the perfect classification effect,there are still a lot of areas to be improved in the study,and this paper only uses several data mining techniques,whether there are other models more suitable for ancient ceramic dating problem,This requires further research.