Application of Weighted Multiple Minimum Support Algorithm Based On FP-Tree in Fault Diagnosis

Abstract

In daily life, device failure is very common, the number and probability of failure is more and more, in order to predict the failure in advance, we need to find the relationship between the failures from the failure data, and accurately predict the failure. Nowadays, most of the methods of data mining and fault prediction use the traditional association rule mining algorithm. However, the traditional association rules mining algorithm uses a fixed minimum support, which can't mine high importance but less frequent itemsets from the data. To solve this problem, someone proposed weighted multiple minimum support algorithm, but it can't use the data mining in the era of big data, and its algorithm efficiency is low. Therefore, this paper proposes weighted multiple minimum support algorithm based on FP Tree for fault prediction, and implements it in Python. Comparing the efficiency of the method with the original method, it is concluded that the algorithm in this paper is more efficient and can be used for fault prediction As a fault prediction method in the era of big data.