Where am I? Using Mobile Sensor Data to Predict a User’s Semantic Place with a Random Forest Algorithm Elisabeth Lex, Oliver Pimas, J¨ org Simon, and Viktoria Pammer-Schindler Know-Center, Graz, Austria Abstract. We use mobile sensor data to predict a mobile phone user’s semantic place, e.g. at home, at work, in a restaurant etc. Such informa- tion can be used to feed context-aware systems, that adapt for instance mobile phone settings like energy saving, connection to Internet, volume of ringtones etc. We consider the task of semantic place prediction as classification problem. In this paper we exploit five feature groups: (i) daily patterns, (ii) weekly patterns, (iii) WLAN information, (iv) battery charging state and (v) accelerometer data. We compare the performance of a Random Forest algorithm and two Support Vector Machines, one with an RBF kernel and one with a Pearson VII function based kernel, on a labelled dataset, and analyse the separate performances of the feature groups as well as promising combinations of feature groups. The win- ning combination of feature groups achieves an accuracy of 0.871 using a Random Forest algorithm on daily patterns and accelerometer data. A detailed analysis reveals that daily patterns are the most discrimi- native feature group for the given semantic place labels. Combining daily patterns with WLAN information, battery charging state or accelerom- eter data further improves the performance. The classifiers using these selected combinations perform better than the classifiers using all feature groups. This is especially encouraging for mobile computing, as fewer fea- tures mean that less computational power is required for classification. 1 Introduction Smartphones currently hold a handheld market share of over 30% - and this market share is rising 1 . Because of their built-in sensors, smartphones are a particularly suitable tool for capturing people’s activities in a physical environ- ment as opposed to people’s interactions with electronic devices or interactions within virtual environments. Such mobile sensor data can be used to analyse behavioural patterns, or within user- and context-adaptive systems. Given the wide spread of smartphones, such systems have the potential to reach an incred- ible amount of users. In this paper, we describe how to use mobile sensor data to predict a mobile phone user’s semantic place, i.e. home, work, restaurant etc. 1 http://mobithinking.com/mobile-marketing-tools/latest-mobile-stats/ a#smartphone-shipments K. Zheng, M. Li, and H. Jiang (Eds.): MOBIQUITOUS 2012, LNICST 120, pp. 64–75, 2013. c  Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2013