Application of PSO-SVM based Text Classification on the Precision Marketing of Insurance

Precision marketing is a new marketing concept put forward by the insurance industry in the field of marketing in recent years. In the context of the big data area, massive data provides a solid scientific basis for precision marketing. Particularly, the text classification algorithm is used to classify consumers' behavior habits and insurance preferences, which provides new possibilities for insurance marketing. However, the accuracy of traditional text classification algorithms is usually not high because of the high text features and sparse features. This article first introduced the particle swarm optimization (PSO) on the support vector machine (SVM) to optimize parameters and select the best parameters to use SVM to classify, and then the accuracy of the algorithm is applied to the HeXun BBS with insurance precision marketing scenarios, and carries on the forecast analysis for the user's consumption behavior, provide a quantitative basis for establishing a targeted marketing strategy, and also to verify the effectiveness of the PSO - SVM on the text classification.