Akilli telefon algilayicilari ve makine öğrenmesi kullanilarak ulaşim türü tespiti Transportation mode detection by using smartphone sensors and machine learning

Bu calismada akilli telefon algilayicilari kullanilarak kullanicilarin ulasim turu tespitinin yapilmasi amaclanmaktadir. Bunun icin kullanicidan yururken, kosarken, bisiklet surerken, araba veya otobus ile seyahat ederken GPS (Global Positioning System), ivmeolcer ve jiroskop algilayicilarindan elde edilen veriler toplanmistir. Veriler 12'ser saniyelik araliklarla etiketlenmis ve toplamda 2500 oruntu elde edilmistir. Bu verilerden 14 oznitelik elde edilmistir. Olusturulan veri seti ile makine ogrenmesi yontemleri kullanilarak testler gerceklestirilmistir. En iyi sonuc GPS, ivmeolcer ve jiroskop algilayicilarinin kombinasyonundan, %99.4 dogruluk orani ile Random Forest yonteminden elde edilmistir. The aim of this study is to detect transportation modes of the users by using smartphone sensors. Therefore, GPS (Global Positioning System), accelerometer and gyroscope sensor data have been collected while walking, running, cycling and travelling by bus or by car from the smartphone of the user. Sensor data were tagged with 12 second interval and 2500 pattern were obtained. 14 features were acquired from the dataset. Machine learning methods were tested on the dataset. Best result was obtained from GPS, accelerometer and gyroscope sensor combination and Random Forest method with 99.4% accuracy rate.

[1]  Gert R. G. Lanckriet,et al.  Identifying Active Travel Behaviors in Challenging Environments Using GPS, Accelerometers, and Machine Learning Algorithms , 2014, Front. Public Health.

[2]  Handan Ankarali,et al.  Random Forests Yöntemi ve Sağlik Alaninda Bir Uygulama , 2011 .

[3]  Hjp Harry Timmermans,et al.  Transportation mode recognition using GPS and accelerometer data , 2013 .

[4]  Sasu Tarkoma,et al.  Accelerometer-based transportation mode detection on smartphones , 2013, SenSys '13.

[5]  Deborah Estrin,et al.  Determining transportation mode on mobile phones , 2008, 2008 12th IEEE International Symposium on Wearable Computers.

[6]  Norbert Gyorbíró,et al.  An Activity Recognition System For Mobile Phones , 2009, Mob. Networks Appl..

[7]  Hao Xia,et al.  Using Smart Phone Sensors to Detect Transportation Modes , 2014, Sensors.

[8]  Hanghang Tong,et al.  Activity recognition with smartphone sensors , 2014 .

[9]  Philip S. Yu,et al.  Transportation mode detection using mobile phones and GIS information , 2011, GIS.

[10]  Hjp Harry Timmermans,et al.  Comparative evaluation of algorithms for GPS data imputation , 2013 .

[11]  Xing Xie,et al.  Learning transportation mode from raw gps data for geographic applications on the web , 2008, WWW.

[12]  Stephen R. Garner,et al.  WEKA: The Waikato Environment for Knowledge Analysis , 1996 .

[13]  Daniel G. Aliaga,et al.  Urban sensing: Using smartphones for transportation mode classification , 2015, Comput. Environ. Urban Syst..

[14]  Miguel A. Labrador,et al.  Centinela: A human activity recognition system based on acceleration and vital sign data , 2012, Pervasive Mob. Comput..

[15]  M. P. Gupta,et al.  Robust Approach for Estimating Probabilities in Naive-Bayes Classifier , 2007, PReMI.

[16]  Cem Ersoy,et al.  Online Human Activity Recognition on Smart Phones , 2012 .

[17]  A. El-Rabbany.  Introduction to GPS: The Global Positioning System , 2002 .

[18]  Jian Ma,et al.  Accelerometer Based Transportation Mode Recognition on Mobile Phones , 2010, 2010 Asia-Pacific Conference on Wearable Computing Systems.

[19]  Peter Widhalm,et al.  Transport mode detection with realistic Smartphone sensor data , 2012, Proceedings of the 21st International Conference on Pattern Recognition (ICPR2012).

[20]  Emiliano Miluzzo,et al.  A survey of mobile phone sensing , 2010, IEEE Communications Magazine.

[21]  Habil Kalkan,et al.  Classification of physical activities using accelerometer signals , 2012, 2012 20th Signal Processing and Communications Applications Conference (SIU).

[22]  Tao Cheng,et al.  Inferring hybrid transportation modes from sparse GPS data using a moving window SVM classification , 2012, Comput. Environ. Urban Syst..