Practice of Machine Learning Algorithm in Data Mining Field

Based on the development of big data era, the processing speed of various information data is accelerated. From the current social information data processing perspective, the traditional information data processing algorithm is no longer applicable to some large-scale data. The application of data mining technology in all walks of life has a very obvious effect. Machine learning algorithm can use artificial intelligence technology, use a large number of data samples to practice, and then automatically find the operation parameters and models. Based on the past achievements, this paper proposes a three-stage positioning method based on support vector machine and K- nearest neighbor method, which greatly improves the positioning accuracy and speed compared with the previous positioning methods based on machine learning. The actual simulation results show that the proposed scheme combines SVM-based positioning method and K- nearest neighbor method, and makes a compromise between positioning accuracy and time complexity, and achieves good positioning results.