Standardized Detection Method of Power Transmission and Transformation Engineering Quantity List Based on Data Mining

This article proposes a standardized detection method for power transmission and transformation project inventory based on data mining. Introducing data mining clustering techniques to cluster the list data of power transmission and transformation projects, using cosine similarity instead of original distance measurement to reduce the dimensionality of the data and construct a ground dimension standard domain. The support vector machine classification mechanism is used to construct a standardized detection model for power transmission and transformation project lists based on standard domains. The model parameters are optimized through algorithms such as expectation maximization to infer the hidden state of the list data. Score based on different levels of allocation points to achieve quantitative evaluation of the standardization level of power transmission and transformation project lists. The experimental results show that the standardization rate of this design method is as high as 92.2%, and the abnormal data detection rate is 82.3%, proving that it can effectively improve the quality of list data.