跳到主要內容

簡易檢索 / 詳目顯示

回結果列表
研究生： 李佩臻
Pei-Chen Lee
論文名稱： Performance Study on Tree-based Classification Algorithms for Smartphone Indoor/outdoor Detection
指導教授： 孫敏德
Min-Te Sun
口試委員:
學位類別： 碩士
Master
系所名稱： 資訊電機學院 - 資訊工程學系
Department of Computer Science & Information Engineering
論文出版年： 2016
畢業學年度： 104
語文別： 英文
論文頁數： 40
中文關鍵詞： 室內外定位 、機器學習 、手機感測器
外文關鍵詞： indoor/outdoor detection, machine learning, IMU
相關次數： 點閱：14  下載：0 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 近年來，室內外定位偵測發展蓬勃，已經成為現今不可或缺的技術之一，並具有許多潛在的應用。舉例來說，可以透過定位使用者是否有從室外環境進入到室內環境， 以便協助使用者手機自動關閉GPS功能來節省能源，或是讓使用者手機自動切換靜音模式。本論文利用tree-based learning 演算法 (i.e., decision trees, boosting, bagging, and random forest) 來分類室內與室外定位的資料。並使用10-fold cross validation來驗證分類結果，以避免分類結果有overfitting的問題發生。最後比較各個tree-based learning 演算法分類結果的性能，並找出其中最適合分類室內外定位資料的tree-based learning演算法。在我們實驗中，雖然分類結果普遍偏高，但其中以boosting的演算法為最佳。Boosting在室內外定位資料分析有高達99%以上的正確率。

    The indoor/outdoor detection for wireless device has many potential applications. For instance, when a device is detected to enter the indoor environment, it can turn off the GPS chip to save energy. In this thesis, the indoor/outdoor detection is treated as the supervised learning problem. The tree-based learning algorithms (i.e., decision trees, boosting, bagging, and random forest) are built from the training dataset and used to classify test dataset. In addition, the 10-fold cross over is used with the algorithms to mitigate the issue of overfitting. The performance of each algorithm are compared to identify the algorithm most appropriate for the indoor/outdoor detection. Although the final performance of each algorithm seems to be high, boosting provides the best accuracy (99%) for indoor/outdoor detection.

    1 Introduction 1 2 RelatedWork 3 2.1 Outdoor positioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2.2 Indoor positioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2.3 Indoor/Outdoor detection . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 3 Preliminary 7 3.1 Decision trees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 3.2 Boosting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 3.3 Bagging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 3.4 Random forest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 3.5 K-fold cross validation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 4 Comparison of the tree-based algorithms 10 4.1 Dataset preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 4.2 Classi cation algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 4.2.1 Decision tree . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 4.2.2 Boosting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 4.2.3 Bagging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 4.2.4 Random forests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 5 Performance 20 5.1 Data collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 5.2 Creating the training and test dataset . . . . . . . . . . . . . . . . . . . . . 20 5.3 Performance metrics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 5.4 Experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 6 Conclusions 28 Reference 29

    [1] Google play. https://play.google.com/store.
    [2] R. https://www.r-project.org/.
    [3] Androsensor, 2015. https://play.google.com/store/apps/details?id=com. fivasim.androsensor&hl=zh_TW.
    [4] A. Amini, R. M. Vaghe , J. M. de la Garza, and R. M. Buehrer. Improving gps-based vehicle positioning for intelligent transportation systems. In 2014 IEEE Intelligent Vehicles Symposium Proceedings, pages 1023{1029, 2014.
    [5] M. S. Arulampalam, S. Maskell, N. Gordon, and T. Clapp. A tutorial on particle lters for online nonlinear/non-gaussian bayesian tracking. IEEE Transactions on Signal Processing, pages 174-188, 2002.
    [6] Guang-Huan Hu, Jia-Jun Bu, and Chun Chen. A novel bayesian framework for indoor-outdoor image classi cation. In Machine Learning and Cybernetics, 2003 International Conference on, pages 3028{3032 Vol.5, 2003.
    [7] Manikanta Kotaru, Kiran Joshi, Dinesh Bharadia, and Sachin Katti. Spot : Decimeter level localization using wi fi. In Proceedings of the 2015 ACM Conference on Special Interest Group on Data Communication, SIGCOMM '15, pages 269{282, 2015.
    [8] Brett Lantz. Machine Learning with R Patterns, Secind Edition. Packt Publishing,2015.
    [9] J. Li and M. Wu. A positioning algorithm of agps. In 2009 International Conference on Signal Processing Systems, pages 385-388, 2009.
    [10] Xingchuan Liu, Qingshan Man, Henghui Lu, and Xiaokang Lin. Wi- /marg/gps integrated system for seamless mobile positioning. In 2013 IEEE Wireless Communications and Networking Conference (WCNC), pages 2323{2328, 2013.
    [11] A. Mai, Z. Wei, and M. Gao. An access control and positioning security management system based on r d. In 2015 7th International Conference on Intelligent Human-Machine Systems and Cybernetics, volume 2, pages 537{540, 2015.
    [12] Carlos Medina, Jos Carlos Segura, and ngel De la Torre. Ultrasound indoor positioning system based on a low-power wireless sensor network providing sub-centimeter
    accuracy. Sensors, 13, 2013.
    [13] Tom M. Mitchell. Machine Learning. McGraw-Hill, 1997.
    [14] Schmidt G T and Brock L D. General questions on Kalman ltering in navigation systems. M.I.T. Instrumentation Laboratory, 1970.
    [15] Sheng-Cheng Yeh, Wu-Hsiao Hsu, Ming-Yang Su, Ching-Hui Chen, and Ko-Hung Liu. A study on outdoor positioning technology using gps and wi networks. In Networking, Sensing and Control, 2009. ICNSC '09. International Conference on,
    pages 597{601, 2009.
    [16] Jiao Yu, Wei-Shinn Ku, Min-Te Sun, and Hua Lu. An RFID and particle lter-based indoor spatial query evaluation system. In Joint 2013 EDBT/ICDT Conferences, EDBT '13 Proceedings, Genoa, Italy, March 18-22, 2013, pages 263-274, 2013.
    [17] X. Yunqiang, Z. Yanshun, W. Z. Qing, and L. Ming. Research for pedestrian navigation positioning method based on mems sensors. In Control Conference (CCC), 2015 34th Chinese, pages 5315-5318, 2015.

    QR CODE
    :::