A Text Classification Algorithm for Power Equipment Defects Based on Random Forest

Abstract

A short text mining architecture with a unique design is suggested to uncover the worth of short texts in the power text and management of power equipment. A Text Classification Algorithm for Power Equipment Defects (TCA-PED) is proposed in this paper. The brief text mining method is initially outlined, with each module’s operation explained in sequence. An adaptation of the short text mining architecture to practical implementation is then presented, based on the particular features of short texts found in electrical equipment power text and management. The samples of faulty texts are submitted to show the deployment of short text mining in designing and management, based on the architecture with the specifically built modules. This framework is well suited to electrical equipment power text and management activities, as demonstrated by the dataset. The particular design of each component also contributes to the enhancement of the system. Finally, the results show the effectiveness of the proposed model.