论文部分内容阅读
为了解决机器学习中的主观信息缺失问题,提出一种新的面向共享数据的迁移组概率学习机(TGPLM-CD).该方法基于结构风险最小化模型,将源领域所含知识和目标领域的类标签组概率信息,特别是领域间的共享数据纳入学习框架中,实现了源领域和目标领域的知识迁移,在待研究领域数据信息不足的情况下提高了分类精确度.大量数据集上的实验结果验证了所提出方法的有效性.
In order to solve the problem of missing subjective information in machine learning, this paper proposes a new TGPLM-oriented group-oriented probability learning machine (TGPLM-CD). This method is based on the structural risk minimization model, which combines the knowledge contained in the source domain and the target domain Class label group probability information, especially the inter-domain shared data into the learning framework to achieve the transfer of knowledge in the source and target areas, improve the classification accuracy in the case of insufficient data and information to be studied. The experimental results verify the effectiveness of the proposed method.