Analysis and Research on Python-Based Identification of Abnormal Commodities on E-Tailing Platforms

With the rapid development of e-commerce, e-tailing platforms have become the main way for people to shop, but e-tailing platforms are also facing some problems, such as untrue information of commodities, false price labeling, and brushing behavior. This paper analyzes and researches the anomaly identification method of e-commerce goods online retail platform. Based on the relevant commodity data and store data information provided by so-and-so big data industry company, the paper firstly adopts exploratory data analysis method, analyzes the characteristics of the data, and draws out the characteristics and causes of abnormal data. Then an anomaly detection method is proposed based on the characteristics of e-commerce goods. This method includes three steps: dimensional organization, extreme anomaly data detection and suspicious data detection, and fully applies machine learning algorithms such as isolated forests to realize the anomaly detection of e-commerce commodity data. Finally, we use the relevant data information provided by So-and-so Big Data Co. Ltd. to verify the effectiveness of the e-commerce commodity (price, sales volume) e-tailing platform anomaly identification method proposed in this paper, and the experimental results show that this system has a high accuracy rate and efficiency.