A Content-based Intelligent Label Classification System for Warehouse Management

Abstract

With the rapid development of the e-commerce business, smart warehouse management has become a critical part of the supply chain for maintaining an optimized cost in stock logistics. The managing of inventory may require substantial labors and expenses on the order fulfillment and the label classification. This paper proposes a warehouse management system that can provide the real-time categorizing service for labels based on a fastText based word text classification algorithm. In this system, a label classifier is presented to facilitate the process of classifying massive labels based on its content. In this classifier, a Huffman tree data structure have been developed and a layered softmax structure is utilized to reduce the computational complexity. Experimental results show that the proposed method is capable of processing efficient label classification tasks where the management efficiency can be improved for the economic benefits of warehouse management.