Mining and analysis of multiple association rules between the Xining loess collapsibility and physical parameters

Abstract

Abstract Collapsibility determination in loess area is expensive, and it also requires a large amount of experimentation. This paper aims to find the association rules between physical parameters and collapsibility of the loess in Xining through the method of data mining, so to help researchers predict the collapsibility of loess. Related physical parameters of loess collapsibility, collected from 1039 samples, involve 13 potential influence factors. According to Grey Relational Analysis, the key influence factors that lead to collapsing are identified from these potential influence factors. Subsequently, take the key influence factors, δs (coefficient of collapsibility) and δzs (coefficient of collapsibility under overburden pressure) as input items, and use the Apriori algorithm to find multiple association rules between them. Then, through analysing the results of association rules between these key influence factors and collapsibility, the evaluation criteria for collapsibility in this area is proposed, which can be used to simplify the workload of determining collapsibility. Finally, based on these research results, recommendations for projects construction were made to ensure the safety of construction in the area.