Research of Sales Anomaly Detecting and Locating Model Based on Big Data
LIU Ju-jun1, JIANG Lei1, PENG Xiong2, ZHOU Qian1, YANG Xian-sheng1
1(School of Computer Science and Engineering, Hunan University of Science and Technology, Xiangtan 411201, China)2(BBK Commercial Chain Limited by Share Ltd, Xiangtan 411201, China)
Abstract：Nowadays, the competition of traditional retailing is very fierce and the amount of data is huge. So mining anomalies under the big data platform and then making auxiliary decision becomes an effective means for enterprises to improve their competitiveness. At present, most outlier detection can only mine the comparable data. However, sales data are affected by season, holiday and other factors and then lose comparability, and the requirement of management is not only mining anomalies, its ultimate purpose is to realize the practical significance such as locating anomalies and making it clear who is to blame. So the method of anomaly Detecting and Locating of sales data becomes an problem. Therefore, a model of Sales Anomaly Detecting and Locating Based on Big Data is proposed. The model makes the data comparable by using the idea of weight. The anomaly location is realized by establishing probability model after outlier detection from different angles. Due to characteristics of weight and unique anomaly location ,the model is highly recognized by relevant professionals when it is applied to the BBK commercial chain Limited by Share Ltd.
刘菊君,姜磊,彭雄,周倩,杨先圣. 大数据下的销售异常发现与定位模型研究[J]. 小型微型计算机系统, 2019, 40(1): 64-68.
LIU Ju-jun,JIANG Lei,PENG Xiong,ZHOU Qian,YANG Xian-sheng. Research of Sales Anomaly Detecting and Locating Model Based on Big Data. Journal of Chinese Computer Systems, 2019, 40(1): 64-68.
2019, Vol. 40 
