Limits...
Beyond where to how: a machine learning approach for sensing mobility contexts using smartphone sensors.

Guinness RE - Sensors (Basel) (2015)

Bottom Line: Furthermore, after tuning the parameters of RandomForest, performance improved to above 97.5%.As a result, the classifiers can be ranked from lowest to highest complexity (i.e., computational cost) as follows: SVM, ANN, LR, BN, DT, NB, IBk, LWL and KStar.The above results suggest that DT algorithms are excellent candidates for detecting mobility contexts in smartphones, both in terms of performance and computational complexity.

View Article: PubMed Central - PubMed

Affiliation: Finnish Geospatial Research Institute, Geodeetinrinne 2, FI-02430 Masala, Finland. robert.guinness@nls.fi.

ABSTRACT
This paper presents the results of research on the use of smartphone sensors (namely, GPS and accelerometers), geospatial information (points of interest, such as bus stops and train stations) and machine learning (ML) to sense mobility contexts. Our goal is to develop techniques to continuously and automatically detect a smartphone user's mobility activities, including walking, running, driving and using a bus or train, in real-time or near-real-time (<5 s). We investigated a wide range of supervised learning techniques for classification, including decision trees (DT), support vector machines (SVM), naive Bayes classifiers (NB), Bayesian networks (BN), logistic regression (LR), artificial neural networks (ANN) and several instance-based classifiers (KStar, LWLand IBk). Applying ten-fold cross-validation, the best performers in terms of correct classification rate (i.e., recall) were DT (96.5%), BN (90.9%), LWL (95.5%) and KStar (95.6%). In particular, the DT-algorithm RandomForest exhibited the best overall performance. After a feature selection process for a subset of algorithms, the performance was improved slightly. Furthermore, after tuning the parameters of RandomForest, performance improved to above 97.5%. Lastly, we measured the computational complexity of the classifiers, in terms of central processing unit (CPU) time needed for classification, to provide a rough comparison between the algorithms in terms of battery usage requirements. As a result, the classifiers can be ranked from lowest to highest complexity (i.e., computational cost) as follows: SVM, ANN, LR, BN, DT, NB, IBk, LWL and KStar. The instance-based classifiers take considerably more computational time than the non-instance-based classifiers, whereas the slowest non-instance-based classifier (NB) required about five-times the amount of CPU time as the fastest classifier (SVM). The above results suggest that DT algorithms are excellent candidates for detecting mobility contexts in smartphones, both in terms of performance and computational complexity.

No MeSH data available.


Related in: MedlinePlus

Plot of accelVariance vs. speed for all mobility classes.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4481999&req=5

f4-sensors-15-09962: Plot of accelVariance vs. speed for all mobility classes.

Mentions: Lastly, a simple way to obtain a better understanding of the cause of errors is to use two-dimensional plots of the features. Since we have seen above that speed and accelVariance are among the most important features, we created plots to see the distribution of the classes with respect to these two variables. Figure 4 below shows such a plot for the entire dataset. For greater clarity, Figure 5 shows only the classes “walking” and “running”. Recall from Table 7 that 6.2% of the “running” instances were confused with the “walking” class. This is not surprising in light of Figure 5, and we will discuss this error type further in the section below.


Beyond where to how: a machine learning approach for sensing mobility contexts using smartphone sensors.

Guinness RE - Sensors (Basel) (2015)

Plot of accelVariance vs. speed for all mobility classes.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4481999&req=5

f4-sensors-15-09962: Plot of accelVariance vs. speed for all mobility classes.
Mentions: Lastly, a simple way to obtain a better understanding of the cause of errors is to use two-dimensional plots of the features. Since we have seen above that speed and accelVariance are among the most important features, we created plots to see the distribution of the classes with respect to these two variables. Figure 4 below shows such a plot for the entire dataset. For greater clarity, Figure 5 shows only the classes “walking” and “running”. Recall from Table 7 that 6.2% of the “running” instances were confused with the “walking” class. This is not surprising in light of Figure 5, and we will discuss this error type further in the section below.

Bottom Line: Furthermore, after tuning the parameters of RandomForest, performance improved to above 97.5%.As a result, the classifiers can be ranked from lowest to highest complexity (i.e., computational cost) as follows: SVM, ANN, LR, BN, DT, NB, IBk, LWL and KStar.The above results suggest that DT algorithms are excellent candidates for detecting mobility contexts in smartphones, both in terms of performance and computational complexity.

View Article: PubMed Central - PubMed

Affiliation: Finnish Geospatial Research Institute, Geodeetinrinne 2, FI-02430 Masala, Finland. robert.guinness@nls.fi.

ABSTRACT
This paper presents the results of research on the use of smartphone sensors (namely, GPS and accelerometers), geospatial information (points of interest, such as bus stops and train stations) and machine learning (ML) to sense mobility contexts. Our goal is to develop techniques to continuously and automatically detect a smartphone user's mobility activities, including walking, running, driving and using a bus or train, in real-time or near-real-time (<5 s). We investigated a wide range of supervised learning techniques for classification, including decision trees (DT), support vector machines (SVM), naive Bayes classifiers (NB), Bayesian networks (BN), logistic regression (LR), artificial neural networks (ANN) and several instance-based classifiers (KStar, LWLand IBk). Applying ten-fold cross-validation, the best performers in terms of correct classification rate (i.e., recall) were DT (96.5%), BN (90.9%), LWL (95.5%) and KStar (95.6%). In particular, the DT-algorithm RandomForest exhibited the best overall performance. After a feature selection process for a subset of algorithms, the performance was improved slightly. Furthermore, after tuning the parameters of RandomForest, performance improved to above 97.5%. Lastly, we measured the computational complexity of the classifiers, in terms of central processing unit (CPU) time needed for classification, to provide a rough comparison between the algorithms in terms of battery usage requirements. As a result, the classifiers can be ranked from lowest to highest complexity (i.e., computational cost) as follows: SVM, ANN, LR, BN, DT, NB, IBk, LWL and KStar. The instance-based classifiers take considerably more computational time than the non-instance-based classifiers, whereas the slowest non-instance-based classifier (NB) required about five-times the amount of CPU time as the fastest classifier (SVM). The above results suggest that DT algorithms are excellent candidates for detecting mobility contexts in smartphones, both in terms of performance and computational complexity.

No MeSH data available.


Related in: MedlinePlus