Machine learning model perspective analysis method

Abstract

The present invention provides a machine learning model perspective analysis method. The method comprises: collecting error data fed back by users and extracting basic information, and extracting correlated information in the feedback data to generate feature space vectors; calculating the scores of query results, employing an original model and a submodel to perform learning classification of the user query result, and obtaining the classification result, namely evaluation scores; and about each user query, calculating nDCG values of the query values, obtaining an actual sequence according to the machine learning model training result, and obtaining an ideal sequence according to the query results and the user query. The machine learning model perspective analysis method analyzes the submodels in the learning model, filters the submodels with bad classification results, selects submodels with good classification results, performs recombination of the selected submodels and generates a new learning model, and the generated new learning model has higher prediction accuracy.

Description

technical field [0001] The invention relates to a perspective analysis method of a machine learning model, which belongs to the field of Internet search. Background technique [0002] With the rapid development of the Internet, search engines have become an important tool for people to use Internet information resources. With the rise and development of Google, Yahoo!, Bing, Baidu and other search engines, the relevance of query results has attracted more and more attention. The ranking of query results has also become the main index for evaluating search engines. [0003] With the rapid development and wide application of information technology, the Internet has developed vigorously and become the world's largest information resource, which has occupied an important position in people's lives. The Internet has also become an important platform for people to share and interact with each other. It is like finding a needle in a haystack for users to find the information t...

AI Technical Summary

Problems solved by technology

This will greatly reduce the efficiency of the search

Once the learning model is established, it becomes more difficult to modify the model.

Images