基于DBSCAN聚類與Apriori關聯(lián)分析的渠道套利識別研究

打開文本圖片集
中圖分類號:TP311.13 文獻標志碼:A 文章編碼:1672-7274(2025)04-0061-03
Abstract: This article proposes a recognition method based on DBSCAN clustering and Apriori correlation analysis fordetecting single channel arbitrage and channel cooperative arbitrage behaviors.Firstly,the order data is processed through stutering and regular text segmentation to form structured data.Then,the DBSCAN clustering algorithm is used to group similarusers and calculate the weights of channels in the group to identify arbitrage chanels. In addition,based on the Apriori algorithm,assciation analysis is conducted onnetwork users,a distance matrix is constructed,and the distanceat which users may handle businessin thechannel is set.The probabilityof users handling businessin the channel is calculated to determine the cooperative arbitrage channel. Compared with traditional audit methods,this methodoptimizes theaudit process,improves audit efficiencyandaccuracyand ensures the compliance and healthy development of enterprise channels.
Keywords: arbitrage identification; clustering algorithm; DBSCAN; text segmentation; correlation analysis: apriori;auditmethods
傳統(tǒng)審計方法采用人工抽樣調(diào)查,耗費時間和人力。(剩余3490字)