• K. VICTOR RAJAN

Online e-commerce exchanges enable the exchange of products and services across organizational or individual boundaries through internet. Transactions are digitally enabled among organizations and individuals. E-commerce exchanges execute a very large number of sales updates when compared to brick-and-mortar stores. Even a few misinterpreted items can have a significant business impact and result in wrong demand/marketing insights. Multiple vendors might sell similar products with slightly different product name or description. Early detection of anomalies in an automated real-time fashion is an important need for such exchanges to predict demand with higher accuracy. Manufactures and suppliers depend on exchange data to estimate demand for a product. Generating demand insight for a particular type of product is a challenging task since vendors do not use common nomenclature for product names or unique product codes globally. Human error also leads to the sale of same product under anomalous names. Many e-commerce exchanges use domain experts to review the product details and group them together before business intelligence reports are generated. With the use of internet growing, e-commerce exchanges see millions of products flowing in every day. Employing man power for matching of products is time consuming and costly as the volume is huge. In this paper, we describe AI-based anomaly detection approach we developed and evaluated for a large-scale online exchange. Our system detects anomalies both in batch and real-time streaming settings, and the items flagged for manual review are found to be very less after automatic processing.

E-commerce Exchange, Anomaly Detection, Demand Insights

"Anomaly Detection of Products in e-Commerce Exchange using Artificial Intelligence for Demand Insights", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.9, Issue 2, page no.b574-b580, February-2022, Available :http://www.jetir.org/papers/JETIR2202167.pdf

"Anomaly Detection of Products in e-Commerce Exchange using Artificial Intelligence for Demand Insights", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.9, Issue 2, page no. ppb574-b580, February-2022, Available at : http://www.jetir.org/papers/JETIR2202167.pdf