=Paper=
{{Paper
|id=Vol-1323/paper24
|storemode=property
|title=A Simulation Study on Automated Transport Mode Detection in
Near-Real Time using a Neural Network
|pdfUrl=https://ceur-ws.org/Vol-1323/paper24.pdf
|volume=Vol-1323
}}
==A Simulation Study on Automated Transport Mode Detection in
Near-Real Time using a Neural Network==
<pdf width="1500px">https://ceur-ws.org/Vol-1323/paper24.pdf</pdf>
<pre>
   A Simulation Study on Automated Transport Mode Detection in
              Near-Real Time using a Neural Network

       Rahul Deb Das                                     Nicole Ronald                                    Stephan Winter
  rahuld@student.unimelb.                           nicole.ronald@unimelb.                            winter@unimelb.edu.au
          edu.au                                             edu.au

                                        Department of Infrastructure Engineering
                                         The University of Melbourne, Australia


                                                               Abstract

                           Detecting transport modes in near-real time is important
                           for various context-aware location based services and
                           understanding urban dynamics. In this paper we present a
                           simulated study on detecting transport modes in near-real
                           time using a neural network. We have shown how
                           detection accuracy will vary with different temporal
                           window sizes and different combination of modes. Since in
                           urban environment transport modes move slowly due to
                           traffic, considering movement attributes or kinematics
                           alone for mode detection is not sufficient. That is why we
                           investigated how spatial information can improve mode
                           detection accuracy. The model has achieved 82%-95%
                           accuracy using different simulation designs and proves its
                           efficacy over other detection models.

1 Introduction
   Transport mode detection from trajectories has seen growing interest in research over last few years for its
importance in various domains such as context-aware computing, location based services, understanding urban
dynamics, travel demand surveys, traffic monitoring, and travel behaviour analysis. Traditionally travel modes have
been surveyed in questionnaires, enabling also to capture additional knowledge including purpose of trip. Travel
surveys, however, are burdensome, erroneous if made from memory, of low spatial detail, and reach only small
sampling rates. Automation should overcome all these issues.

   Since the late 1990's, due to advancements in positioning and navigation technology, GPS started being used as a
mean to collect travel data and assess its reliability and future possibilities (Wolf, 2000). Eventually, the use of GPS
has increased as it has become more precise, portable and ubiquitous. Nowadays people themselves can track their
movement trajectories using GPS and potentially other sensors on-board their smart phones (Periera et al., 2013).

  Most of the research on travel mode detection is based on rigid velocity based model. However, a velocity based
approach is not always sufficient. For example, low speed conditions, which are nowadays typical in urban traffic
due to traffic at capacity, or bad weather, produce mode ambiguities. In low speed traffic conditions, the speed of a
bus is similar to a car or bicycle. Therefore, there is a need to consider various non-kinematic attributes along with
movement attributes (kinematics) in order to detect different transport modes.


Copyright © by the paper’s authors. Copying permitted only for private and academic purposes.
In: B. Veenendaal and A. Kealy (Eds.): Research@Locate'15, Brisbane, Australia, 10-12 March 2015, published at http://ceur-ws.org




  Research@Locate '15                                               46
   Existing transport mode detection research is mostly offline. That means modes are detected once a trip is
completed from historical trajectories. Existing methods use the entire trip record in the form of trajectory and then
separates the trajectory based on walking based segments into number of meaningful parts that correspond to
respective transport modes.

   The hypothesis behind this research is that a neural network based model can adjust well in real time with
varying movement behaviour and overcomes the mode ambiguity under low speed conditions.

   In detecting transport modes, movement characteristics derived from trajectories of the users are the raw data
source. In this paper we will concentrate on trajectories of a single sensor, as provided by GPS enabled smart
phones. Such GPS trajectories are unlabelled and come in raw format. Other sensors in the phone are neglected for
the time being, but can easily be included in the model. A classifier is required that can detect the various transport
modes used along each trip in real to near-real time. In this paper, a neural network based classifier has been tested.
Our contributions are as follows-

   1) We developed a simulated near-real time transport mode detection model based on a multi-layer perceptron
neural network.

   2) Earlier approaches to neural network based transport mode detection are mostly offline and did not use any
spatial information. In this research we show how spatial information can increase the detection accuracy.

   3) Selecting a proper temporal window for detecting transport modes in near-real time is critical and context
dependent. In this paper we investigate how detection accuracy varies with different temporal window sizes, which
helps in selecting a proper window size based on accuracy requirement.

   In this paper we also evaluate the performance measure of a multi-layer perceptron neural network in order to
detect transport modes in near-real time. A real time model can detect the transport mode epoch by epoch basis
(such as second by second). In this research, we simulated queries within short temporal window to detect a given
mode instead of second by second basis. Hence, we call this model a near-real time mode detection approach.

   Detecting transport mode in near-real time is comparatively an emerging research area. In this paper we have
developed a basic but intuitive near-real time mode detection model using a supervised learning approach. Real
time mode detection can be useful for a number of applications. Applications include various context-aware
location based services where the context could be a given transport mode. A petrol pump can distribute an
electronic discounted coupon within its neighbourhood to all the private cars only. Detecting transport modes in real
time can also help developing various context-aware mobile applications that can sense the modality and act
accordingly. One instant could be developing a mode-dependent auto-answering service on smart-phones. If the
mobile senses the owner is in driving mode then the auto-answer can automatically be enabled and helps driver to
concentrate on the road rather than receiving any incoming call. Thereby this can help in reducing distractions on
the road in order to reduce road accidents. This approach can also be helpful for urban planners or emergency
service providers who want to know people’s mode choice at a given route or in a given region at a given time
window for modeling travel demand or various spatio-temporal events.

   The paper is organized as follows. Section 2 discusses related works in transport mode detection from various
perspectives. Section 3 discusses some of the basic terminologies and methodology. Section 4 demonstrates data
preparation and experimentation. Section 5 shows the experimental results. Section 6 presents the discussion of
these results, and Section 7 concludes the paper.

2 Related work
    Nowadays, smart-phones come with GPS enabled facilities. Since smart-phones are carried by the users almost
everywhere and all the time, hence, this positioning facility can be utilized in order to collect trajectories without
any external intervention. Once a GPS trajectory has been collected or is in the process of being collected, those
trajectories or part there-of can be used for transport mode detection in real time or post-processing mode. In this
regard, existing work is mostly based on post-processing of the trajectories or detecting modes offline. Existing
literature shows a wide variety of post-processing algorithms and classifiers. Some of the approaches used the
classification technique directly without segmenting the GPS trajectories (Byon et al., 2007; Dodge et al., 2009;
Reddy et al., 2010). At the same time there are approaches applying segmentation of the entire trajectory into
meaningful parts, corresponding to different modes, before classification (Mountain and Raper, 2001; Tsui and
Shalaby, 2006; Schussler and Axhaussen, 2009; Zheng et al., 2010; Biljecki et al., 2012; Hemminki et al., 2013).

   Segmentation is done based on those points that show high probability of mode change. Mountain and Raper
used change in speed and direction for segmentation in their work (Mountain and Raper, 2001). However, this
approach creates ambiguities in certain cases where the vehicles move slowly and constrained to specific roads or




 Research@Locate '15                                      47
the rail or tram networks. Liao et al. used proximity to potential change points, such as bus stops or train stops for
offline mode detection (Liao et al., 2007). However, GPS accuracy greatly varies in urban environments, depending
on the number of satellites in view, time of the day and season, atmospheric conditions and surrounding sources of
multipath effects. Other research used change in peaks of acceleration curves in order to segment the trajectory
(Hemminki et al., 2013). However this approach also suffers from low ambiguity resolution, typically in low speed
condition such as during bad weather or traffic congestion. Another common and intuitive way for segmenting the
trajectory is based on detecting walking segments. The rationale behind this approach is the observation that a
person generally walks between using two modes of transport. This approach has achieved promising results for
segmentation (Tsui and Shalaby, 2006; Zheng et al., 2010; Biljecki et al., 2012). However this approach also fails
when there is a quick mode change or walking is negligible.

   There have been a number of different algorithms for mode classification used so far. Zheng and colleagues used
a decision tree, Bayesian Net, Conditional Random Field (CRF) and Support Vector Machine (SVM) in their work
with 75% reported accuracy (Zheng et al., 2010). Gonzalez et al. used neural networks with 91% accuracy
(Gonzalez et al., 2010). Some works are solely based on statistical measures (Patterson et al., 2003).

   As far as the input parameters or indicators are concerned, prior work mostly concentrated on velocity attributes
(Bohte et al., 2008; Schussler and Axhausen, 2009). But in low speed condition velocity and acceleration are not
sufficient to resolve the ambiguities. So, more recently, research has incorporated additional movement attributes
including heading rate change and stop rates (Zheng et al., 2010). Vibration data has also been tested as an
additional attribute with promising results (Ohashi et al., 2013). However, in order to achieve better accuracy and
account for GPS signal loss others have used inertial localization and navigation sensors such as accelerometers,
along with GPS sensors (Reddy et al., 2010; Hemminki et al., 2013; Ohashi et al., 2013).

   Byon and colleagues used GPS trajectories collected by GPS loggers to study detection accuracy in real time.
However there focus was mainly on how accuracy varies with different sampling frequencies (Byon et al., 2009).
They achieved high detection accuracy at 20 min temporal window. However they observed mainly four modes
auto, walk, car, bus. Although Byon and colleagues developed two neural network models, one route specific and
another one a universal model, they did not explore how spatial knowledge can help in detecting different modes.
Also their approach is limited by their use of GPS loggers: they used instantaneous speed, acceleration, number of
satellites in view for a given transport mode to train their classifier. Number of satellites in view depends on
particular transport mode. Such as GPS device inside a bus is obstructed by the metallic body and ceiling and
vertical windows limiting the number of satellites in view. Whereas a car would have wider front windshield that
would allow stronger and multiple GPS signals. However when using smart-phones for detecting modes,
instantaneous acceleration, number of satellites in view or horizontal dilution of precision values may not be
available.

   Gonzalez and colleagues developed a neural network based mode detection model with a core focus on how to
reduce streaming of movement data. Earlier work used a static and fixed data transmission procedure but that
suffered from high financial costs associated with data transmission as well as computational overhead and storage
issues. Gonzalez and colleagues proposed a novel critical point (CP) algorithm to transmit only the relevant GPS
points during the trip (Gonzalez et al., 2010).

   Since movement states are uncertain and imprecise there are a couple of mode detection appraoch using fuzzy
logic (Tsui and Shalaby, 2006; Biljecki et al., 2010). A fuzzy approach with three criteria and five to ten modes has
been tested with an accuracy of more than 90% (Schussler and Axhausen, 2009; Biljecki et al., 2010). However
these approaches are rule-based and involve fuzzy antecedents and fuzzy consequents (Zadeh, 1965; Mamdani and
Assilian, 1975). This approach cannot adapt with different movement behaviour in real time. Since fuzzy logic
based models are developed based on expert knowledge with predefined premise and consequents hence they are
not scalable with new parameters and thus pose scalability and flexibility issues. In this paper we present a neural
network based model that can learn in real time. A neural network based model is flexible and scalable.

3 Theory

In this section we presented some basic definitions, concepts and methodology used in this research.

3.1 Raw Trajectory
   A raw trajectory is a set of spatio-temporal points arranged in a chronological order. This can be mathematically
expressed as
   Tr ={Pi }: Pi= (xi, yi, zi, ti) ; i є [0, N] ;∀ i : (ti<ti+1) ……………………………………………. (1)




 Research@Locate '15                                     48
3.2 Segment
    Any connected part of a raw trajectory with a specific semantics is a segment. For example, if a part of trajectory
is extracted with a given annotated mode, then that is a modal segment. Similarly, if certain part(s) of a trajectory is
extracted over a given time period that part would be a temporal segment.
3.3 Model architecture
   The architecture of a multi-layer feed forward back propagation neural network is explained as follows-

   A multi-layer feed forward neural network consists of mainly three layers- a) input layer, b) hidden layer, c)
output layer. These layers contain one or more than one nodes or neurons. Input nodes are connected to hidden
nodes and hidden nodes are connected to output nodes. But nodes of the same layer have to be disjoint and they
cannot be connected to each other. Input layer is responsible to get input signals from the external world typically in
the form of movement attributes (kinematics) or spatial attributes (non-kinematics) in the context of transport mode
detection.

3.4 Training
   Neural network can learn online and adapt well with given instances. However before using a neural network, it
has to be trained to map a given set of inputs to a given output class. The training typically starts from input layer as
soon as input stimuli are fed in. Nodes in each layer receive input signal from the preceding layer and send an
output signal to the nodes in immediate succeeding layer. Each node multiplies the input signal with a previously
established weight, adds a threshold, converts into an output signal through an activation function and sends it to the
other nodes in the succeeding layer. The hidden layer is not directly connected to real world. This is the most
important layer that processes the information and creates categorizing features for classification which sends signal
to the output layer to categorize a given set of feature vectors. Once the output signal produces a response it is
evaluated with the actual response. The difference between the predicted response and desired response is the error
term of the neural network which is then back propagated to the model in order to adjust the weight and threshold
values iteratively. This iterative process goes on in a cyclic way until a prescribed number of cycles (epochs) or a
desired error level is achieved during training phase.

  The rate at which a neural network learns can be adjusted by changing certain parameters called learning rate
(LR) and momentum (M). These parameters control the change in weight and their persistence throughout total
number of epochs.

3.5 Near-real time simulation
    In order to detect transport modes in near-real time, queries will be fetched to a central server with kinematic
and non-kinematic information. In this research we used a set of historical trajectories for near-real time simulation
purpose. In order to train a neural network model small temporal segments have been extracted from the
trajectories. Kinematic and non-kinematic attributes are then estimated over that temporal window in order to
capture various movement behaviour of the given mode within that time period. Since a transport mode can exhibit
different movement behaviour at different instant hence there is a need to train the classifier with movement
behaviours for each mode at different instant of time over different trajectories. In order to extract movement
behaviour of each transport mode, temporal segment over a given temporal window of a given mode segment has
been extracted at regular interval of time.
3.6 Mode segmentation
   Each trajectory can be expressed as a set of modal segments. This can be expressed as
   T={SMj} ………………………………………………………………………………………… (2)
   Where, j ϵ [1, N]; N= total number of modes used by the user over the trajectory
   Each modal segment can be expressed as SMj= {Pij, Mij}………………………………………. (3)
   where i= ith spatio-temporal index; j= jth modal segment index; SMj= jth segment of the trajectory; Pij= ith patio-
temporal point in jth segment; Mij= mode for ith point in jth segment.
   A modal segment can be divided into a number of overlapping temporal segments. That can be mathematically
expressed as -
   SMj={TWkj}t : k ϵ [1, M] …………………………………………………………………..….. (4)
   Where, TWkj= kth temporal segment of j modal segment over time window t; M= total number of temporal
segments over a given modal segment.
3.7 Temporal segmentation
   Once a trajectory is segmented into number of modal segments then each modal segment is segmented in
number of temporal segments overlapping by (n-1) spatio-temporal points, where n is the number of spatio-
temporal points in a given temporal segment. The overlap is chosen as n-1 in order to capture diverse movement
behaviour of the given mode at a finer granularity.




 Research@Locate '15                                       49
Each temporal segment can be expressed as TWkjt={Pijk, Mijk} ……………………………. (5)

   Where, TWkjt= kth temporal segment of ‘t’ time length in jth modal segment
   Pijk= ith spatio-temporal point in kth temporal segment of jth modal segment
   Mijk= Annotated mode in kth temporal segment of jth modal segment
3.8 Kinematics and spatial information
   In this research eight kinematic attributes are estimated using Euclidean functions in space-time domain such as
average speed, average acceleration, variance of speed, variance of acceleration, maximum speed, maximum
acceleration, minimum speed and minimum acceleration. In order to understand how a given mode behaves
spatially with respect to different spatial objects (route network or POI), eight spatial relevance measures have been
considered (see Table 2). Spatial relevance with respect to different spatial objects is calculated based on spatial
proximity of spatio-temporal points to the given spatial object or a part thereof.

3.9 POI relevance estimation
   In order to estimate POI relevance (proximity to bus stop, train stop, traffic signal or car wash or parking lot) a
density-based clustering kernel is ran over each temporal window. Then POI relevance over a given temporal
segment is estimated as

   POIRelc=POIRelc-1+ s*(n/N) ……………………………………………………….…….. (6)
   Where, POIRelc= Relevance measure of a given POI over cluster ‘c’ over temporal window [t1, t2]
   POIRelc-1= Relevance measure of a given POI over cluster ‘c-1’ over temporal window [t1, t2]
   s= scaling factor (s=10 in this case)
   n= number of elements in the cluster falling in the search radius of the given POI
   N= total number of the elements in the cluster
3.10 Instance formation
   In order to train and test the model using N-fold cross-validation, instances are created in the form of feature
vectors which include kinematics and spatial attributes estimated over each temporal segment and fed into the
model. A flowchart is given to show the workflow in Figure 1.

3.11 Performance measure
In order to evaluate the performance of the model, we used N-fold cross-validation. Since in N-fold cross-validation
all the feature space is used using N-1 as training and 1 set of feature vectors as test thus it can capture the state
behaviour at a fine granularity. However in hold-back type training, the accuracy of the model depends on the
percentage of training instances that can represent all the details and characteristic behaviour of the entire
population. Since in real time mode detection instances may vary with a temporal window size and modal
movement behaviour, hence performance measure of a N-fold cross-validation strategy has been presented in this
research (see experiment and result section), assuming an iterative N-fold cross-validation over growing time can
dynamically improve the model in near-real time.




                       Figure 1: Flowchart of simulated mode detection modeling architecture




 Research@Locate '15                                       50
4 Data preparation and experimental setup
   In order to evaluate our hypothesis Microsoft’s GeoLife datset has been used in this experiment (Zheng et al.,
2008a; Zheng et al., 2008b; Zheng et al., 2010). The dataset mainly covers Beijing CBD and its surrounding suburb.
The dataset was collected by smart-phones and GPS loggers in the form of GPS trajectories. Different sampling
intervals ranging from 2-5 seconds have been used in this dataset. Users provided their trajectories and ground truth
separately. This dataset contains various transport modes such as car, walk, bus, taxi, train, subway and bike.
However, no accuracy measure such as horizontal dilution of precision (HDOP) is provided in this dataset. The
dataset also suffers from semantic gaps due to both technical reasons such as urban canyons, indoor environments
and misreporting such as time gap between annotating two different mode segments by a user. During pre-
processing stage some of the GPS points are found to be outside the study area. There were also some
inconsistencies in the annotations, such as walking over unreasonable longer duration or with unreasonable speed.
There are also semantic gaps during signal loss and missing annotations in the dataset. The portions of the dataset
containing such semantic inconsistencies are discarded. However a future work can use signal loss and contextual
information to detect certain modes such as train in a subway.




Figure 2: Beijing GeoLife GPS dataset overlaid on the road network (in blue colour) and rail network (in red
colour)

   In a data filtration stage 2.5 m/s has been set as walking speed threshold (Minetti, 2000), and some of the
trajectories are discarded. In this experiment 264 trajectories have been used including training and test trajectories
(Fig. 2).

   Earlier works used HDOP value and can easily filter noise points (Byon et al., 2009; Gonzalez et al., 2010), but
in this dataset we do not have any information that can provide positional accuracy or confidence level for each
GPS fix. Hence, we setup two different experimental designs. One with filtered GPS data points where walking
speed more than 2.5 m/s have been removed. Another setup was used without any filtering walking speeds. In both
cases, raw velocity values are smoothed using an inverse distance weightage (IDW) smoothing kernel.

   However technically a segment can also be viewed as a trajectory if it is treated discretely for further analysis. In
order to detect transport modes in near-real time, we used a portion of a historical GPS dataset with transport mode
annotated to train the mode detection model. Then we generated queries randomly at different instant of the
trajectory and fetch the queries to the model to detect the transport mode as if the queries are coming in near-real
time. A multi-layer perceptron (MLP) neural network has been realized in this research in order to detect various
transport modes. The reason neural network has been investigated in this research because neural network is
flexible, easily scalable and most importantly, it can learn online and adapt well.

    In this research, five transport modes are considered: car, walk, bus, train and bike. Since car and taxi are
difficult to distinguish especially in near-real time hence car and taxi are both grouped as car for time being.
However in future car and taxi can be treated separately depending on the availability of contextual information.
Similarly, train or light rail and subway are grouped as train. A multi-layer perceptron (MLP) neural network has
been modelled using Weka, a Java based open source machine learning package. Since the time window is a critical
factor in near-real time mode detection hence different temporal window size has been evaluated such as 120 sec,
180 sec, 240 sec, 300 sec and 600 sec based on subjective judgement. Experiments are also set up using only
kinematic information and spatial information along with kinematics. For kinematic information eight movement
attributes over a given temporal window have been considered since different modalities may exhibit different
movement behaviour (Table 1). When kinematic attributes are used a 8-6-5 MLP was formed, and using spatial and
kinematic attributes a 16-10-5 MLP model was used to detect different transport modes (Fig. 3). A popular



 Research@Locate '15                                      51
approach to select number of hidden nodes can be calculated as the closest integer value of [(input nodes+target
nodes)/2]. Hence we selected 6 hidden nodes when input nodes are 8 and output nodes are 5. Likewise, for 16 input
nodes and 5 target nodes, number of hidden nodes are 10.

Table 1: Kinematic attributes

Attribute                                                 Relevance
Average speed (avgSpeed)                                  Central tendency of a temporal segment in order to
Average acceleration (avgAccl)                            approximate a characteristic movement behaviour
Variance of speed (varSpeed)                              Spread of movement behaviour over the temporal
Variance of acceleration (varAccl)                        window
Maximum speed (maxSpeed)                                  Upper bound of respective movement attributes within
Maximum acceleration (maxAccl)                            a given temporal window
Minimum speed (minSpeed)                                  Lower bound of respective movement attributes within
Minimum acceleration (minAccl)                            a given temporal window

For spatial information eight spatial attributes including proximity to route network and different POIs over a given
temporal window have been considered (Table 2).

Table 2: Spatial attributes

Attribute                                                 Relevance
Average road proximity (avgRoadProx)                      Central tendency of proximity distribution over a
Average railway proximity (avgTrainProx)                  given temporal window
Variance of road proximity (varRoadProx)                  Spread of proximity distribution over a given temporal
Variance of railway proximity (varTrainProx)              window
Relevance score for bus stop (busRel)                     Relevance measure of each relevant cluster from a
Relevance score for train stop (trainRel)                 given POI based on spatial proximity
Relevance score for traffic stop (trafficRel)
Relevance for parking lot and car wash (plcwRel)

In order to study the performance of neural network through different training strategies, the model has been
realized through N-fold cross-validation (where N=10). Table 3 shows number of instances used in N-fold cross-
validation for different time window.

Table 3: Time window vs instances

Time window                                               Instances
120                                                       15060
180                                                       13735
240                                                       12600
300                                                       11488
480                                                       9166
600                                                       7835


5 Experimental results
    The performance of the model has been evaluated on five modes against different temporal window size using
filtered walking speeds in order to compare with the existing works that used positional uncertainty information. An
experiment has also been carried out without filtering walking speeds assuming in real time people may run instead
of walking during mode transfer or GPS positions can be subjected to various errors leading to walking speed
greater than any threshold value.

   The model was trained and tested using 10-fold cross-validation. In the first stage accuracy was tested against
different temporal window sizes on trajectories where walking speeds are filtered. In the second stage accuracy was
evaluated without filtering the trajectories in order to simulate real time mode detection. In both cases, the model
shows that using spatial information can easily outperform the accuracy produced by only kinematics attributes.
The reason behind this is that all transport modes may move slowly during traffic congestion or bad weather which
leads to mode ambiguities. Figure 4 shows how mode ambiguity may arise using only acceleration measure. In this
figure different modes may not be distinguished from their acceleration since they are clustered around the similar
acceleration measures (Fig. 4).




 Research@Locate '15                                     52
Figure 3: A 16-10-5 MLP neural network based mode detection model. L1 indicates input layer, L2 indicates
hidden layer and L3 indicates output layer.

   But when spatial relevance, in particular proximity to route network or given POI relevance, is considered the
mode has been detected more accurately. In figure 5, the bus mode shows high bus stop relevance and hence bus
mode is more prominent from other modes. However since walking can take place anywhere over the footpath near
the bus stop hence, some of the walking instances have shown high bus stop relevance owing to false positives (Fig.
5). The rational is, a car can travel like a train with similar speed and acceleration but the underlying route network
would be different and POI relevance will also vary accordingly.




Figure 4: Mode ambiguities from similar acceleration distribution. Class map: {blue: walk; red: car; green: bus;
indigo: train; brown: bike}. X-axis: modes; Y-axis: acceleration value

   From the accuracy measures it is clear that there is a trade-off between temporal window size and mode
detection accuracy. Selecting an optimal window size is context dependent. Overall 300 sec seem to an optimal
window size for near-real time mode detection as the accuracy starts increasing gradually from this point and the
accuracy measure is more than 82% for unfiltered trajectories and 86% for filtered trajectories (Fig. 6; Fig. 7).




Figure 5: Bus stop relevance vs different modes. Class map: {blue: walk; red: car; green: train; indigo: bus; brown:
bike}; X-axis: modes and Y-axis: bus relevance



 Research@Locate '15                                      53
Figure 6: Accuracy measure with filtered walking speed            Figure 7: Accuracy without filtered walking speed

    The methodology was also tested on car, walk and train assuming these modes will show quite distinct behaviour
in terms spatial relevance as well as kinematic relevance. When spatial information was used accuracy reached 95%
for filtered trajectories and 93% for unfiltered trajectories (Fig. 9). A spatial visualization is also presented to show
the classification accuracy for three modes (train, car and walk). The diagonals are true positives and off-diagonals
are false positives (Fig. 8). The figure shows the model can give a high accuracy and less type I and type II error for
walking. However due to similar kinematic behaviour some of the car instances are mostly classified as walk owing
to type I error. Likewise train instances are sometimes classified as walk and car.

   In order to compare state-of-the-art approaches that used only kinematic information, another test was conducted
within temporal window of 300 sec, on car, bus and walk modes. It was found there was a small difference in
estimated accuracy by using only kinematics inputs, and kinematics and spatial inputs together. However using
spatial information and kinematics, the accuracy is certainly more than that of using kinematics alone. The small
difference of accuracy can be justified as the bus network has not been used in this research; only the road network
was used. A car or bus both can travel on road network and hence it was not easy to distinguish between car and
bus. But there is a good chance that car and bus can be easily distinguished by using a bus network. Using spatial
information average accuracy for car, walk and bus was achieved 81.24 % whereas without spatial information the
accuracy was 79.50 %.

   We also compared the performance of our MLP neural network with some of the well-studied machine learning
algorithms. The result shows a MLP neural network outperforms other approach. Interestingly the accuracy of a
MLP neural network increases as the size of the time window increases whereas other approaches show saturation
over growing time window. This clearly shows the ability of a MLP to learn and adapt well in near-real time as
more instances come in with fine and varied state behaviour of different modes (Table 4).




Figure 8: Classification accuracy visualization. Class map: {blue: walk; red: car; green: train}; X-axis indicates
actual modes and Y-axis indicates predicted modes



6 Discussions
From the result it is evident that spatial information can improve mode detection accuracy significantly, especially
in near-real time. In near-real time detecting different transport modes is challenging from their movement
attributes only, as the queries are issued for a very short interval and different modes may have similar movement
behaviour. Earlier literature did not consider minimum speed and acceleration as all of them are offline and based
on segmenting the entire trajectory in each modal segment that normally starts with zero speed and zero
acceleration (Gonzalez et al., 2010).




 Research@Locate '15                                       54
                                    Figure 9: Accuracy measure for three modes

Table 4: Accuracy measures for different classifiers against different time window

Time window (sec)      SVM (%)                MLP Neural             Logistic   regression   RBF Network (%)
                                              network (%)            (%)
120                    59.10                  74.08                  68.51                   58.77
180                    60.65                  79.65                  70.32                   60.37
240                    64.15                  79.34                  72.51                   64.15
300                    61.78                  82.55                  75.20                   66.89
480                    66.84                  86.63                  77.86                   68.95
600                    68.35                  87.05                  77.76                   66.87


   But in real time since a query can come at any time and the given mode need not to start or finish from stationary
state within the given temporal window, hence minimum and maximum bounds, central tendency and variance of
speed and acceleration distribution over the given time window have been considered. In order to supplement the
mode detection accuracy various spatial information have been considered assuming the fact that each mode shows
distinct spatial behaviour, such as bus and car will travel along the roadway; likewise train will travel on train
network only. However, in this research the bus network of Beijing was not used, instead the entire road network
consisting of the bus network and roads together was used. In order to distinguish different modes especially bike,
car and bus different POI relevance has been considered such as bus stop for bus, traffic for bike, bus and car, car
wash and parking lot for car. However, it has been observed that when the time window is short it cannot capture
distinct behaviour and thus leads to ambiguities especially for car, bus, walk and bike that share similar movement
and spatial relevance at some instants.

   From individual classification accuracy for five modes it has been observed walking has the highest false
positives as different modes can be slowed down and behave like walking. At the same time this can also give false
negative as people can walk on the road, near the bus stop, traffic signal or parking lot and hence respective POI
relevance may be higher for a walking segment. That gives a false impression of other modes corresponding to the
respective spatial relevance.

   There is also a trade-off between the temporal window and the accuracy that raises questions of selecting the
proper time window for a given location based service. Say for emergency services, the amount of response time
required is less than the time required by an urban planner or traffic engineer to understand travel demand from
people’s mode choice or location based context-aware advertisements. This also poses challenges of selecting a
proper and optimal window size to detect transport modes as accurately as possible. However from this study it is
evident that the window size bears an inverse relationship with the accuracy measure, assuming the mode is not
changed within the temporal window.

   Earlier work using neural network only performed offline analysis considering only three modes. The highest
reported accuracy achieved was 91% (Gonzalez et al., 2010). However our approach capable of real time estimation
that shows accuracy can reach up to 95% or more when using the road network along with other relevant
information for three modes, and 87% when using five modes over 600 sec. However the accuracy depends on
number of modes, their spatial relevance and the size of the temporal window. The accuracy also depends on the
clustering algorithm used to estimate POI relevance over a given temporal window.




 Research@Locate '15                                     55
   However, at this moment this approach is limited by detecting only a single mode within a given temporal
window. But in real time there is a possibility that a quick transfer can take place from one mode to another mode
followed by walking. This creates composite mode segments within a same temporal window which is difficult to
detect using only GPS signals (Das et al., 2014). In order to distinguish two different modes with in a composite
segment different inertial sensor information along with GPS is required that can distinctly detect presence of two
different modal class within a given temporal segment from their characteristic kinematic signatures.

7 Conclusions
   Detecting transport modes in near-real time is an emerging research area. This is particularly useful in various
context-aware location based services and understanding urban dynamics in near-real time. In order to detect
various transport modes Microsoft’s GeoLife dataset has been used in this research. In this research a simulated
near-real time mode detection classification framework has been developed using a neural network based classifier.
We have evaluated the performance of neural network in detecting various modes, since neural networks can adjust
well with different input and output parameters online. Neural networks also offer flexibility and scalability in
terms of learning ability and accommodating new information from the external world. In this paper, we
particularly focused on how real time mode detection accuracy varies with varying temporal window size. This has
been figured out within a small temporal window all the transport modes show similar kinematic behaviour. In
order to detect different modes more accurately we used various spatial information such as route network
information and POI information.

   We tested our hypothesis on three sets of modes: two sets containing three modes and one set containing five
modes). Our result shows incorporating spatial information can improve mode detection accuracy. We achieved
accuracy 95% accuracy on three modes only and 93% accuracy on all five modes. The result also shows a MLP
neural network can outperform other machine learning algorithms with growing temporal window size.

   Future work will look into distinguishing a composite segment within a given temporal window where a quick
transfer has occurred. In order to explore different modes within a temporal segment, different sensor signals such
as accelerometer, proximity sensor, gyroscope information are required that can give characteristic movement
behaviour of each modes at a very fine granularity. In this research while forming the clusters we only considered
spatial relevance of each cluster with respect to given POI. We did not consider temporal relevance as temporal
window may vary from as small as 120 sec to as high as 600 sec or more. During smaller temporal window, it is
difficult to set a temporal relevance or dwell time. Future research will address spatio-temporal issues while
developing potential clusters within a given temporal segment to calculate spatio-temporal relevance for each mode.

References

Biljecki, F., Ledoux, H., & Oosterom, P. V. (2012). Transportation mode-based segmentation and classification of
   movement trajectories. International Journal of Geographical Information Science, 17(2), 385-407.

Bohte, W., Maat, K., & Quak, W. (2008). A method for deriving trip destinations and modes for GPS-based travel
  surveys. In J. V. Schaick & S. V. d. Spek (Eds.), Urbanism on Track : IOS Press, pp. 129-145.

Byon, Y., Abdulhai, B., Shalaby, A. S. (2007). Impact of sampling rate of GPS enabled cell phones on mode
  detection and GIS map matching performance. Transportation Research Board Annual Meeting 7(1795), 21.

Byon, Y., Abdulhai, B., & Shalaby, A. (2009). Real-time transportation mode detection via tracking Global
  Positioning System mobile devices. Journal of Intelligent Transportation Systems: Technology, Planning, and
  Operations,, 13(4), 161-170.

Das., R. D., Ronald., N., & Winter., S. (2014). Clustering based transfer detection with fuzzy activity recognition
  from smart-phone GPS trajectories, 17th IEEE International Conference on Intelligent Transportation Systems,
  Qingdao, China.

Dodge, S., Weibel, R., & Forootan, E. (2009). Revealing the physics of movement: Comparing the similarity of
  movement characteristics of different types of moving objects. Computers, Environment and Urban Systems,
  33(6), 419-434.

Gonzalez, P., Weinstein, J., Barbeau, S., Labrador, M., Winters, P., Georggi, N., & Perez, R. (2010). Automating
  mode detection for travel behaviour analysis by using Global Positioning Systems enabled mobile phones and
  neural networks. Intelligent Transport Systems, IET, 4(1), 37-49.




 Research@Locate '15                                    56
Hemminki, S., Nurmi, P., & Tarkoma, S. (2013). Accelerometer-based transportation mode detection on
  smartphones. 11th ACM Conference on Embedded Networked Sensor Systems, Rome, Italy.

Hemminki, S., Nurmi, P., & Tarkoma, S. (2013). Accelerometer-based transportation mode detection on
  smartphones. 11th ACM Conference on Embedded Networked Sensor Systems, Roma, Italy.

Liao, L., Fox, D., & Kautz, H. (2007). Extracting places and activities from GPS traces using hierarchical
   conditional random fields. International Journal of Robotics Research, 26(1), 119-134.

Mamdani, E.H. and Assilian, S, (1975). An experiment in linguistic synthesis with a fuzzy logic controller,
  International Journal of Man-Machine Studies, 7(1), 1-13.

Minetti, A. (2000). The three modes of terrestrial locomotion. Biomechanics and Biology of Movement B. M. Nigg,
  B. R. MacIntosh, and J. Mester, Eds., ed: Human Kinetics, pp. 67–78.

Mountain, D., & Raper, J. (2001). Modelling human spatio-temporal behaviour: a challenge for location-based
  services. GeoComputation, Brisbane, Australia.

Ohashi, H., Akiyama, T., Yamamoto, M., & Sato, A. (2013). Modality classification method based on the model of
  vibration generation while vehicles are running. 6th ACM SIGSPATIAL International Workshop on
  Computational Transportation Science, Orlando, USA.

Patterson, D., Liao, L., Fox, D., & Kautz, H. (2003). Inferring high-level behavior from low-level sensors. Lecture
   Notes in Computer Science, Berlin, Germany.

Pereira, F., Carrion, C., Zhao, F., Cottrill, C. D., Zegras, C., & Ben-Akiva, M. (2013). The Future Mobility Survey:
   Overview and preliminary evaluation. Proceedings of the Eastern Asia Society for Transportation Studies.

Reddy, S., Mun, M., Burke, J., Estrin, D., Hansen, M., & Srivastava, M. (2010). Using mobile phones to determine
  transportation modes. ACM Transactions on Sensor Networks, 6(2), 1-27.

Schuessler, N., & Axhausen, K. (2009). Processing GPS raw data without additional information. 88th Annual
  Conference of the Transportation Research Board, Washington D.C. USA.

Tsui, S. Y. A., & Shalaby, A. S. (2006). Enhanced system for link and mode identification for personal travel
  surveys based on Global Positioning Systems. Transportation Research Record, 1972, 38-45.

Wolf, J. (2000). Using GPS data loggers to replace travel diaries in the collection of travel data. PhD thesis, Georgia
 Institute of Technology, Atlanta, USA.

Zadeh, L. A. (1965). Fuzzy sets, Information and Control 8(3), 338-353.

Zheng, Y., Chen, Y., Li, Q., Xie, X., & Ma, W.-Y. (2010). Understanding transportation modes based on GPS data
  for Web applications. ACM Transactions on The Web, New York, USA.

Zheng, Y., Li, Q., Chen, Y., & Xie, X. (2008). Understanding mobility based on GPS data. ACM conference on
  Ubiquitous Computing (UbiComp 2008), Seoul, Korea. ACM Press: 312–321.

Zheng, Y., Liu, L.,Wang, L., & Xie, X. (2008). Learning transportation modes from raw GPS data for
  geographic application on the web, International conference on World Wide Web (WWW 2008), Beijing, China.
  ACM Press: 247-256.




 Research@Locate '15                                      57

</pre>