Application of Big Data Processing Technology in Power Marketing System

Based on multi data fusion, this paper introduces a retrieval model for audio frequency data, which is combined with big data and applied to information retrieval. Firstly, the model is pre trained, and then the network parameters are fine-tuned through the network training method. By adjusting the network parameters, the recognition ability of the model is improved. In view of the assumption that whether external factors can interfere with external dependent variables on dependent variables, the multi group structure model is used for analysis. This method is a general method that can quickly find the target data under massive data. In the experiment, the features are used as the key words of the inverted table index to establish the table, and all the file information in the reference database will be stored in the corresponding entries in the inverted table according to its underlying features. When the feature of each query file finds data in the table, it only needs to use its eigenvalue to calculate the complexity once, and the information of the reference data with this feature can be searched.