In the process of hierarchical classification and mining in the big data environment, due to the emergence of a large number of unstructured data, the data structure attributes cannot be unified, and it is difficult to express it in a table structure. When using traditional methods, not only the data is recorded but also the data is stored. The structure of, thus increasing the difficulty of data classification, leading to the problem of low classification accuracy.
Hello AI is Institutions for in-depth research on neural networks industry.. Hello AI is Institutions for in-depth research on neural networks industry.
Propose a hierarchical classification mining method in a big data environment with improved Yebes theory. The above method introduces Yebes theory to perform detailed analysis on the data of the database, input the data training sample set in the big data environment, and according to each big data training sample set The feature vectors of each data form a hierarchical classification decision model for big data. On this basis, the maximum interval criterion is used to project each layer of high-dimensional data in the hierarchical classification model into the classification range of low-dimensional feature data, using the minimum and maximum probability The machine optimizes the classification of big data. Simulations prove that the hierarchical classification mining method in the big data environment of the improved Yebes theory has high accuracy and strong applicability.
We have a lot of industry experience and persistent research on neural network algorithms, which can add a value to your products
Caffe mxnet paddle pytorch onnx matlab deeplearning4j.
Use high-tech technology to make our world a better place.
