A Method for Satisfaction Evaluation of Search Engine Users' Information Needs by Fusion of Multi-view and Semi-Supervised Learning

Abstract

The invention relates to a search engine user information demand satisfaction evaluation method that combines multi-view and semi-supervised learning. The user satisfaction model of supervised learning and six stages of evaluation, the present invention uses a small amount of labeled data and a large amount of unlabeled data to improve the performance of the evaluation model through the method of semi-supervised learning, and introduces the idea of ​​multi-view learning to overcome the traditional Semi-supervised learning methods of views are prone to fall into the problem of local optima. The beneficial effects are: (1) the user satisfaction of search engine information needs can be effectively evaluated with a small amount of labeled data; (2) the user satisfaction model evaluation can be improved by using a small amount of labeled data and a large amount of unlabeled data performance; (3) Describe the user's search process from the perspective of behavior and time, and avoid the model from falling into local optimum through mutual learning.

Description

technical field [0001] The invention relates to the field of Internet information technology, in particular to a search engine user information demand satisfaction evaluation method integrating multi-view and semi-supervised learning. Background technique [0002] With the rapid development of knowledge economy and informatization construction, the scale of network information data is rapidly expanding. Massive information resources not only enrich people's information sources, but also cause troubles for people to obtain information. Search engines, with their increasingly precise and humanized Information retrieval service has become one of the main tools for users to access the World Wide Web to find and obtain resource information. At the same time, search engines need to continuously improve their algorithms and optimize their systems to meet the increasing information needs of users and the requirements for efficient and convenient access to information resources. The...

AI Technical Summary

Problems solved by technology

[0003] Traditional search engine quality evaluation indicators such as precision at n (Precision at n, P@n), average accuracy (Mean Average Precision, MAP), normalized discounted cumulative return (normalize Discounted Cumulative Gain, nDCG), etc. need to use a large number of Manually labeling data to evaluate the performance of search engines, but this kind of manual labeling needs to consume a lot of manpower and time resources, it is difficult to carry out large-scale real-time

Semi-supervised learning can make the evaluation method automatically use a large amount of unlabeled data to assist in the learning of a small amount of labeled data. However, most traditional semi-supervised learning methods are based on single-view, that is, simply combine all sub-attribute sets in the data into one A single attribute set ignores the unique statistical properties of each sub-attribute, and it is easy to fall into local optimum when the training data is extremely scarce.

Images