Heterogeneous transfer learning for activity recognition using heuristic search techniques. Issue 4 (28th October 2014)

• Record Type: Journal Article
• Authors: Dillon Feuz, Kyle; J. Cook, Diane
• Is Part Of: International journal of pervasive computing and communications. Volume 10:Issue 4(2014)
• Page Start: 393
• Page End: 418
• Publication Date: 2014-10-28
• Subjects: Heterogeneous transfer learning -- Domain adaption -- Genetic algorithms -- Activity recognition -- Heuristic search -- Smart environments; Ubiquitous computing -- Periodicals; Mobile computing -- Periodicals; Computer network protocols -- Periodicals; Computer network architectures -- Periodicals; Application software -- Development -- Periodicals; 004.6
• DOI: 10.1108/IJPCC-03-2014-0020
• Languages: English
• ISSNs: 1742-7371

Abstract

Abstract : Purpose: – The purpose of this paper is to study heterogeneous transfer learning for activity recognition using heuristic search techniques. Many pervasive computing applications require information about the activities currently being performed, but activity recognition algorithms typically require substantial amounts of labeled training data for each setting. One solution to this problem is to leverage transfer learning techniques to reuse available labeled data in new situations. Design/methodology/approach: – This paper introduces three novel heterogeneous transfer learning techniques that reverse the typical transfer model and map the target feature space to the source feature space and apply them to activity recognition in a smart apartment. This paper evaluates the techniques on data from 18 different smart apartments located in an assisted-care facility and compares the results against several baselines. Findings: – The three transfer learning techniques are all able to outperform the baseline comparisons in several situations. Furthermore, the techniques are successfully used in an ensemble approach to achieve even higher levels of accuracy. Originality/value: – The techniques in this paper represent a considerable step forward in heterogeneous transfer learning by removing the need to rely on instance – instance or feature – feature co-occurrence data.