DOI: 10.3724/SP.J.1004.2013.02186

Acta Automatica Sinica (自动化学报) 2013/39:12 PP.2186-2192

Self-adaptive Transfer for Decision Trees Based on Similarity Metric

Negative transfer, transfer opportunity and transfer method are the most key problems affecting the learning performance of transfer learning. In order to solve these problems, a self-adaptive transfer for decision trees based on a similarity metric (STDT) is proposed. At first, according to whether the source task datasets to be allowed to access, a prediction probability based on constituents or paths is adaptively used to calculate the a±nity coe±cient between decision trees, which can quantify the similarity degree of related tasks. Secondly, a judgment condition of multi-sources is used to determine whether the multi-source integrated transfer is adopted. If do, the similarity degrees are normalized, which can be viewed as transfer weights assigned to source decision trees to be transferred. At last, the source decision trees are transferred to assist the target task in making decisions. Simulation results on UCI and text classification datasets illustrate that, compared with multi-source transfer algorithms, i. e., weighted sum rule (WSR) and MS-TrAdaBoost, the proposed STDT has a faster transfer speed with the assurance of high decision accuracy.

Key words:Transfer learning, decision tree, similarity metric, affinity coefficient

ReleaseDate:2014-07-21 17:04:32

Funds:National Natural Science Foundation of China (61072094, 61273143), Special Grade of the Financial Support from China Postdoctoral Science Foundation (20110095110016, 20120095110025), and College Graduate Research and Innovation Projects of Jiangsu Province (CXZZ12 0932)