Research On Landslide Hazard Data Mining Based On Association Rule

Landslides will have adverse effects on the safety of the ecological environment and the safety of human life and property,and in order to avoid risks,it is necessary to predict landslide disasters in time.Data mining is the process of extracting useful information from a large number of data,due to the long-term monitoring of landslides,a large number of landslide disasters related to data have been obtained,so the use of data mining technology for landslide disasters can save a certain amount of manpower and material resources on the basis of landslide disaster prediction,for landslide disaster prediction provides a new idea.Among them,the Apriori algorithm in the association rule is one of the commonly used data mining algorithms,and its principle is simple and easy to implement,which is very suitable for studying the correlation between the factors causing landslide disasters.However,the traditional Apriori correlation algorithm not only scans the database multiple times during the process,but also generates a large number of unnecessary candidate sets,so the algorithm is less efficient in the face of a large amount of data.In view of the shortcomings of the algorithm,an improved algorithm based on the classical Apriori algorithm is proposed,and a landslide disaster data mining system is developed based on this improved algorithm,and the specific work is as follows:Firstly,in order to improve the computational efficiency of the classical Apriori algorithm,an improved algorithm based on the Apriori algorithm is proposed and named the SM-Apriori(Simplified Matrix Apriori)association rule algorithm.The algorithm reduces the number of database scans by introducing a Boolean matrix,which solves the problem of multiple database scans during the operation of the classical Apriori algorithm.On this basis,a new method of simplifying the Boolean matrix is proposed,and the generation of candidate sets is reduced by using the reduced matrix with the nature of Apriori algorithm,which solves the problem of generating a large number of unnecessary candidate sets during the operation of classical Apriori algorithm.In addition,weight vectors are introduced to simplify the calculation of support,further reducing the running time of the algorithm.Simulation experiments show that the SM-Apriori algorithm has significantly improved efficiency compared with the classical Apriori algorithm and the FP-Growth algorithm.Secondly,the paper adopts the object-oriented design method,researches and develops a data mining system based on SM-Apriori algorithm based on MVC architecture,which uses PyQt5 Designer development tools for user interface design on the Python platform,and develops a C/S data information management system with local data files as the data management system,which includes a total of five subsystems: data files,data processing,data visualization,model operation and result analysis.Therefore,the information management functions such as data import,preprocessing,and visual analysis are realized,and finally the SM-Apriori algorithm is introduced into the system on the basis of the above work,and the function of using the SM-Apriori algorithm to mine landslide disaster related data is realized.In order to realize the purpose of analyzing and mining the relevant data of landslide disasters,this paper proposes a more efficient SM-Apriori algorithm based on the classical Apriori algorithm,and develops the corresponding data mining system based on the algorithm,and the final frequent itemsets and mining results are displayed in the system,which successfully achieves the purpose of finding out the hidden association rules by mining landslide disaster-related data.The SM-Apriori algorithm obtains the frequent itemset by introducing the matrix,simplifying it,and then searching the candidate set layer by layer,and then calculates the confidence and improvement between the items in the final frequent itemset,and finally generates the association rule.By analyzing,interpreting and explaining the generated association rules,the more accurate the rule prediction with strong correlation,and finally achieve the purpose of disaster prediction.