Method and device for generating classification model

Abstract

The invention discloses a method and device for generating a classification model, and relates to the technical field of computers. A specific embodiment of the method comprises the following steps: acquiring a first data set for training; constructing a feature wide table according to the first data set; based on a pre-constructed knowledge graph, determining a division proportion corresponding to the feature wide table according to a current target scene and the data scale of the feature wide table, so as to divide the feature wide table into a training set and a test set; determining a classification algorithm and one or more general parameters corresponding to the classification algorithm based on the knowledge graph according to the current target scene and the current classification problem; and according to the training set and the test set, performing training by using the classification algorithm to generate a classification model. According to the embodiment, data set division and classification algorithm and classification algorithm general parameter selection can be automatically carried out based on the knowledge graph, and the classification model generation efficiency is improved.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a method and device for generating a classification model. Background technique [0002] With the rise and development of artificial intelligence, machine learning has become the focus and hotspot of research, while classification problems such as binary classification and multi-classification problems have become a basic and important part of machine learning due to their wide application prospects. [0003] In order to solve the classification problem, algorithm engineers often need to spend a lot of energy writing programs for training classification models based on the existing algorithm framework, and the programs for training classification models written cannot be reused, which greatly reduces the generation efficiency of classification models. In addition, algorithm engineers need to select the classification algorithm and divide the data set based on personal past expe...

AI Technical Summary

Problems solved by technology

[0003] In order to solve the classification problem, algorithm engineers often need to spend a lot of energy writing programs for training classification models based on the existing algorithm framework, and the programs for training classification models written cannot be reused, which greatly reduces the generation efficiency of classification models

In addition, algorithm engineers need to select the classification algorithm and divide the data set based on personal past experience before writing the program for training the classification model, which wastes a lot of energy, and due to the limitations of personal past experience, it cannot be based on the selected classification algorithm. And after dividing the data set to obtain a better classification model

Images