Sentiment classification method based on multi-source field instance migration

Abstract

The invention relates to a sentiment classification method based on multi-source field instance migration. For the phenomenon that the migration efficiency of a TrAdaBoost algorithm in migration learning may be lowered, multi-source learning is introduced, by trying sample migration from different source fields or combining the features of multiple source fields, migration learning is stable and effective, much initial weight is distributed to a target sample, so that weight mismatching is relieved, and the phenomenon of quote imbalance is improved by resampling data at each step; and for the phenomenon of source field weight premature convergence in the TrAdaBoost algorithm, dynamic factors are added, and the problem that weight entropy is transferred from a source sample to the target sample is solved. According to the method, premature convergence of the weight of the source field sample which is small in the correlation with the target field is avoided, a learning target task is helped together, and knowledge of all source fields is fully used.

Description

technical field [0001] The invention relates to a sentiment classification method based on multi-source field example migration. Background technique [0002] According to the similarity between different tasks, transfer learning transfers the source domain data to the target domain, realizes the utilization of existing knowledge, makes the traditional learning from scratch become accumulative learning, and improves the learning efficiency. The characteristic is to use knowledge in related fields to help complete the learning tasks in the target field. There are many ways to express relevant knowledge in the source domain and the target domain, which can be divided into sample instances, feature maps, model parameters, and association rules. Choosing an appropriate transfer learning method for different knowledge representation methods is a prerequisite for ensuring learning in the target domain. [0003] For the study of knowledge expressed as sample instances, the focus i...

AI Technical Summary

Problems solved by technology

[0008] Step 1), introduce multi-source learning, by transferring samples from different source domains or combining the characteristics of multiple source domains, transfer learning is more stable and effective, assigning more initial weights to target samples, and alleviating weight mismatch The problem, and resampling the data at each step to improve the phenomenon of reference imbalance;

Images