An Improved Hidden Markov Model Based on Weighted Observation
WANG Changhai1, LI Zhehui2, WANG Bo1, XU Yuwei3, HUANG Wanwei1
1.Software Engineering College, Zhengzhou University of Light Industry, Zhengzhou 450002;
2.Henan Provincial Institute of Scientific and Technical Information, Zhengzhou 450003;
3.School of Cyber Science and Engineering, Southeast University, Nanjing 211189
As the classic hidden Markov model(HMM) loses the sight of confidence of labeled results while building a sequence, a weighted observation hidden Markov model(WOHMM) is proposed. The algorithms in the steps of probability calculation, parameter learning as well as sequence labeling are described in detail. The simulation results on the public datasets show that the parameters obtained by the parameter learning algorithm of WOHMM are closer to the real values than those of HMM, and the performance of sequence labeling algorithm is superior to the state-of-the-art methods.
王昌海, 李哲辉, 王博, 许昱玮, 黄万伟. 基于加权观测的隐马尔可夫模型[J]. 模式识别与人工智能, 2019, 32(6): 515-523.
WANG Changhai, LI Zhehui, WANG Bo, XU Yuwei, HUANG Wanwei. An Improved Hidden Markov Model Based on Weighted Observation. , 2019, 32(6): 515-523.
[1] BULLING A, BLANKE U, SCHIELE B. A Tutorial on Human Activity Recognition Using Body-Worn Inertial Sensors. ACM Computing Surveys, 2014, 46(3). DOI: 10.1145/2499621.
[2] SCHEURER S, TEDESCO S, BROWN K N, et al. Human Activity Recognition for Emergency First Responders via Body-Worn Inertial Sensors // Proc of the 14th IEEE International Conference on Wearable and Implantable Body Sensor Networks. Washington, USA: IEEE, 2017: 5-8.
[3] SANCHEZ V G, PFEIFFER C F, SKEIE N O. A Review of Smart House Analysis Methods for Assisting Older People Living Alone. Journal of Sensor and Actuator Networks, 2017, 6(3). DOI: 10.3390/jsan6030011.
[4] YANG Z, WU C S, ZHOU Z, et al. Mobility Increases Localizability: A Survey on Wireless Indoor Localization Using Inertial Sensors. ACM Computing Surveys, 2015, 47(3). DOI: 10.1145/2676430.
[5] HAMMERLA N Y, PLOETZ T. Let′s (not) Stick Together: Pairwise Similarity Biases Cross-Validation in Activity Recognition // Proc of the ACM International Joint Conference on Pervasive and Ubiquitous Computing. New York, USA: ACM, 2015: 1041-1051.
[6] WANG Z L, WU D H, GRAVINA R, et al. Kernel Fusion Based Extreme Learning Machine for Cross-Location Activity Recognition. Information Fusion, 2017, 37: 1-9.
[7] HOSEINI-TABATABAEI S A, GLUHAK A, TAFAZOLLI R. A Survey on Smartphone-Based Systems for Opportunistic User Context Recognition. ACM Computing Surveys, 2013, 45(3). DOI: 10.1145/2480741.2480744.
[8] PHAN T. Improving Activity Recognition via Automatic Decision Tree Pruning // Proc of the ACM International Joint Conference on Pervasive and Ubiquitous Computing. New York, USA: ACM, 2014: 827-832.
[9] KRISHNAN N C, PANCHANATHAN S. Analysis of Low Resolution Accelerometer Data for Continuous Human Activity Recognition // Proc of the IEEE International Conference on Acoustics, Speech and Signal Processing. Washington, USA: IEEE, 2008: 3337-3340.
[10] ZHENG L X, WU D H, RUAN X Y, et al. A Novel Energy-Efficient Approach for Human Activity Recognition. Sensors, 2017, 17(9). DOI: 10.3390/s17092064.
[11] ZHONG M Y, WEN J H, HU P Z, et al. Advancing Android Activity Recognition Service with Markov Smoother // Proc of the IEEE International Conference on Pervasive Computing and Communication Workshops. Washington, USA: IEEE, 2015: 38-43.
[12] ZHONG M Y, WEN J H, HU P Z, et al. Advancing Android Activity Recognition Service with Markov Smoother: Practical Solutions. Pervasive and Mobile Computing, 2017, 38: 60-76.
[13] YOUNES R, MARTIN T L, JONES M. Activity Classification at a Higher Level: What to Do After the Classifier Does Its Best? // Proc of the ACM International Symposium on Wearable Computers. New York, USA: ACM, 2015: 83-86.
[14] WANG C H, XU Y W, LIANG H, et al. WOODY: A Post-Process Method for Smartphone-Based Activity Recognition. IEEE Access, 2018, 6: 49611-49625.
[15] RABINER L, JUANG B. An Introduction to Hidden Markov Mo-dels. IEEE Assp Magazine, 1986, 3(3): 4-16.
[16] WEN J H, ZHONG M Y, INDULSKA J. Creating General Model for Activity Recognition with Minimum Labelled Data // Proc of the ACM International Symposium on Wearable Computers. New York, USA: ACM, 2015: 87-90.
[17] WEN J H, WANG Z Y. Learning General Model for Activity Re-cognition with Limited Labelled Data. Expert Systems with Applications, 2017, 74(15): 19-28.
[18] KIM Y J, KANG B N, KIM D. Hidden Markov Model Ensemble for Activity Recognition Using Tri-axis Accelerometer // Proc of the IEEE International Conference on Systems, Man, and Cybernetics. Washington, USA: IEEE, 2015: 3036-3041.
[19] 汪成亮,王小均.基于三轴传感器的老年人日常活动识别.电子学报, 2017, 45(3): 570-576.
(WANG C L, WANG X J. Daily Activity Recognition Based on Triaxial Accelerometer of Elderly People. Acta Electronica Sinica, 2017, 45(3): 570-576.)
[20] 王昌海,张建忠,徐敬东,等.基于HMM的动作识别结果可信度计算方法.通信学报, 2016, 37(5): 143-151.
(WANG C H, ZHANG J Z, XU J D, et al. Identifying the Confidence Level of Activity Recognition via HMM. Journal on Communications, 2016, 37(5): 143-151.)
[21] WANG C H, XU Y W, ZHANG J Z, et al. SW-HMM: a Method for Evaluating Confidence of Smartphone-Based Activity Recognition // Proc of the International Symposium on Parallel and Distributed Processing with Applications. Washington, USA: IEEE, 2016: 2086-2091.
2019, Vol. 32 
