SlideShare a Scribd company logo
1 of 35
Download to read offline
ENTER 2015 Research Track Slide Number 1
Transportation Mode Annotation of Tourist GPS
Trajectories under Environmental Constraints
Hidekazu Kasahara, Mikihiko Mori,
Masayuki Mukunoki, and Michihiko Minoh
Kyoto University, Japan
Hidekazu.kasahara@mm.media.kyoto-u.ac.jp
http://www.mm.media.kyoto-u.ac.jp/en/members/hidekazu-
kasahara
ENTER 2015 Research Track Slide Number 2
Agenda
• Motivation
• Purpose
• Previous Work
• Problems
• Proposed Method
• Experiment & Evaluation
• Discussion
• Future Work & Conclusion
ENTER 2015 Research Track Slide Number 3
Motivation
• DMOs want to know the tourists’ activities.
– Policy making / destination marketing strategy.
• Optimum deployment of transportation.
• Outdoor ads / direction boards.
Spot B
Spot A
Bus
1,000 tourists
Taxi
10 tourists
Ads on bus is more effective
than those on taxi in this route.
Accurate transportation usage
statistics is necessary for DMOs
for the decision making.
The traditional stats is costly.
ENTER 2015 Research Track Slide Number 4
Purpose of Study
• Annotating transportation mode from tourists’
GPS trajectory (x, y, t).
• Modeling the tourists activity as a chain of
transportation modes.
• Transportation usage statistics is calculated from
these chains of transportation modes.
• In addition, tourists’ personal preferences can be
explored from the chains.
Walk 60 min. Bus 20 min.Bus 20 min.
WalkWalk
5 min.
Train 20 min.
ENTER 2015 Research Track Slide Number 5
Scope of Study
• Tourists move by foot and public transportation.
– Train, route bus and taxi as public transportation.
– Tour bus, bicycle, motor-bike and hire car are out of
scope.
• Destinations that include numerous locations of
interest, such as Milano.
• Only GPS.
– Acceleration meter is out of scope.
ENTER 2015 Research Track Slide Number 6
Previous Work
• Approaches using tourist moving physical features.
– Speed: Decision tree (Aoki 2008, Zheng 2010).
– Baysean network (Stenneth 2011).
– Hierarchical CRF (Liao 2007).
– SVM (Bolbol 2012).
– Acceleration: Adaptive boosting (Hemminki 2013).
– Regression analysis (Furukawa 2014).
• Approach using environmental factors.
– Assumption on typical tourist travel patterns (Stopher 2008).
– Bus stops and road network (Liao 2007).
– Land utilization situation (Yan 2013).
ENTER 2015 Research Track Slide Number 7
Transportation Mode Estimation
using Tourist Speed
Yellow points shows
deceleration at bus stops
and crossings.
Station
Temple
Castle
Train
Golden
Pagoda
Yellow: Under 1km/hour
Black : 1-60 km/hour
Red : Over 60 km/hour
These decelerations are
estimated as walk / retain
modes using by speed.
GPS Trajectory
1km
ENTER 2015 Research Track Slide Number 8
Annotation using
Environmental Constraints
bus bus train train train bus
p1
p2 p3
p4
p5 p6
Estimate using Tourist Features
(Speed)
• In physical space, there are
some environmental constraints
restrict transportation usage.
– Environmental constraints.
ENTER 2015 Research Track Slide Number 9
Annotation using
Environmental Constraints
Environmental Constraints :
All 6 Points Are on a Bus Route
Not on a railway.
Estimate using Tourist Features
(Speed)
• In physical space, there are
some environmental constraints
restrict transportation usage.
– Environmental constraints.
bus bus train train train bus
p1
p2 p3
p4
p5 p6
ENTER 2015 Research Track Slide Number 10
Annotation using
Environmental Constraints
Environmental Constraints :
All 6 Points Are on a Bus Route
Not on a railway.
Estimate using Tourist Features
(Speed)
Inconsistency
• In physical space, there are
some environmental constraints
restrict transportation usage.
– Environmental constraints.
• In the case when the estimation
with tourist moving features is
failed, there is regional
inconsistency with
environmental constraints.
bus bus train train train bus
p1
p2 p3
p4
p5 p6
ENTER 2015 Research Track Slide Number 11
Problem
• The system failed to estimate car and train
as retain or walk modes when they
decelerate or stop in case the system use
only tourist features (speed).
• Examples of deceleration:
– Route bus decelerates at all bus stop along the
route.
– Train decelerates at stations on the railways.
ENTER 2015 Research Track Slide Number 12
Proposed Method
• Using the environmental constraints,
decrease the inconsistency between the
estimate using tourist features (speed) and
environmental constraints.
• Annotate the transportation modes with
the least inconsistency in total.
• We evaluate the proposed methods
compared with human labeling.
ENTER 2015 Research Track Slide Number 13
Proposed Method
Output : GPS Trajectory Consists of (x, y, time, speed)
• Environmental
constraints
• Speed
Output: Fragmented Segments
Output: Merged Segments with the least inconsistency in total
• Interleave
• Continuity
• Filtering
• Speed calculation
Technique
‫ݏ‬ଵ ‫ݏ‬ଶ ‫ݏ‬ଷ ‫ݏ‬ସ ‫ݏ‬ହ
݉ଵ ݉ଶ ݉ଵ ݉ଶ ݉ଵ
‫ݏ‬ଵ ܵଶ
ᇱ
‫ݏ‬ହ
݉ଵ ݉ଶ ݉ଵ
Preprocess
Stage 1
Stage 2
ENTER 2015 Research Track Slide Number 14
Preprocess & Stage #1
(x, y, time) : GPS raw data
(x, y, time, speed)
Filtering & speed calculation
On Train Route
On Bus Route
On Motorway
On Pedestrian area
(x, y, time, speed, mode, segment)
speed > sp_train
Speed Environmental Constraints
sp_walk < speed
speed > sp_walk
and
and
… Stop frequency
ENTER 2015 Research Track Slide Number 15
#2 Stage: Segment Merger
• Interleave Assumption:
Tourists do not change
“train to car” or
“car to train” directly.
• Continuity Assumption:
Tourists typically do not
change the transportation
mode over a short period.
Bus
Taxi
Walk
Bus
Never Directly
Connected
Too short to change
ENTER 2015 Research Track Slide Number 16
Experiment
• Dataset
– 16 persons’ GPS trajectory data from students
and teachers who travelled.
• Destination : Kyoto
• 160,130 points in total.
• GPS measures location per 1 second.
• From 8:00 to 18:00.
ENTER 2015 Research Track Slide Number 17
Screening Value
• sp_walk = 2 metres per second.
– The screening value between walk and bus/taxi.
• sp_train = 15 metres per second.
– The screening value between bus/taxi and train.
• ݀݁݊‫݁ݏ‬ = 10%
– The screening value of stop frequency between bus
and taxi. (Route bus stops more frequent than taxi.)
• Interleave = 240 seconds
– If the length of segment is under 240 seconds, the
segment is merged to next or before segment
ENTER 2015 Research Track Slide Number 18
Environmental Constraints
• Data is supplied as shape files.
Bus
Route
(Gov. data)
Train
Route
(COTS)
Motorway
(COTS)
Pedestrian
Zone
(Gov. data)
ENTER 2015 Research Track Slide Number 19
Results of the Experiment
Bus Taxi Train Walk
Total
count
Bus
82.8%
(15,601)
17.5%
(1,413)
0.0%
(0)
4.5%
(5,817) (22,831)
Taxi
6.1%
(1,158)
60.1%
(4,847)
10.6%
(529)
0.3%
(371)
(6,905)
Train
0.2%
(30)
0.3%
(21)
77.3%
(3,869)
1.3%
(1,620)
(5,540)
Walk
10.9%
(2,048)
22.1%
(1,779)
12.2%
(609)
93.9%
(120,418)
(124,854)
Total
count
100.0%
(18,837)
100.0%
(8,060)
100.0%
(5,007)
100.0%
(128,226)
90.4%
(160,130)
Estimate
Actual
ENTER 2015 Research Track Slide Number 20
Comparison
Bus Taxi Train Walk
Total
Proposed 82.8% 60.1% 77.3% 93.9% 90.4%
SVM
*1) 88.9 % 78.9% 67.1% 67.7% 75.2%
Speed without
environmental
constraints
*2)*3)
44.3% 2.5% 49.2% 99.1% 86.2%
Method
Mode
*1) Result of 5-fold cross validation on the same data set
*2) Bus route constraints are used for bus / taxi estimation.
*3) The universe is biased to “walk” mode.
ENTER 2015 Research Track Slide Number 21
Discussion
• Automatic estimation
– The screening values are given by human in the
proposed method.
• Generalization ability is low.
• These values should be decided automatically.
– Stochastic function.
• Function can be estimated
from training data.
• Stochastic function varies
in different environments 0%
20%
40%
60%
80%
100%
Mode Probability Function
Walk Tran Taxi Bus
ENTER 2015 Research Track Slide Number 22
Discussion
• Context among primitive components
– Transportation modes consist of three primitive
components; “stable speed move”, “acceleration
(deceleration)” and “stop.”
– We call context to the relationship between
consecutive segments of each primitive component.
– We should consider the context for estimation.
– We have two assumptions of context in this study and
give some screening values. However, these should be
automatically decided by using training data.
ENTER 2015 Research Track Slide Number 23
Discussion
• Distinction of bus and taxi is difficult
– Only using speed, this system cannot distinguish
between taxi and bus accurately.
• Other environmental constraints is necessary.
– Other environmental constraints is necessary.
• Other environmental constraints
– Traffic confusion.
– Bus stop location.
ENTER 2015 Research Track Slide Number 24
Conclusion & Future Work
• We proposed a new GPS semantic annotation method
using environmental constraints based on two
assumptions of tourist behaviour.
• The results indicate the high accuracy, 90%.
• Making model based on stochastic logic.
– Improving generalization capability of the method.
• By using other environmental constraints, we try to
improve the performance in future.
ENTER 2015 Research Track Slide Number 25
Appendix
ENTER 2015 Research Track Slide Number 26
Transportation Mode Annotation
using Tourist Features
Annotation using
tourist’s speed
is valid for retain /
walk modes.
Golden Pagoda
Blue : 10-60km/hour
Yellow: Under 1km/hour
Green : 1-10 km/hour
In this area, system
estimated retain /
walk / bus modes.
GPS Trajectory
ENTER 2015 Research Track Slide Number 27
Transportation Mode Estimation
from GPS Trajectory
Walk 60 min. Bus 20 min.
Chain of Transportation
Modes Bus 20 min.
Bus
GPS Trajectory Walk
Temple
ENTER 2015 Research Track Slide Number 28
What’s school trip?
School trip is one of the biggest group tours in Japan
• School trip is important for DMOs and travel agents
• 10% of all stayed tourists in Kyoto 2012 is school trip
students
• The number of students who participated school trip in
2012 is 3.4 million
• Participation rate of students is high
• 94.4% of junior high schools /75.5% of senior high schools
Unit : million
ENTER 2015 Research Track Slide Number 29
Escort-teachers
Field HQ (HOTEL)
School staff
School(Home)
School master
Real-time monitoring
Current position
Smartphone
(Group leader carries)
Trajectory
Tablet/PC
- GPS & Wi-Fi positioning
Overview of safety ensuring system
ETSS (Educational Tour Support System)
Student groups
Information Sharing among All Related Persons
Group Leader(Student. Trained before trip.)
- Mail & Voice
Safety ensuring
No Navigation
For education
System
Tablet
ENTER 2015 Research Track Slide Number 30
Screenshot of student phone
Schoolmaster
Homeroom
teacher
Voice & MailDisaster MapNormal Map
School NameNow 1 hr All
EvacuationArea
ObservingAttractions
Current Position Evacuation Area MapNormal MapNormal Map
ENTER 2015 Research Track Slide Number 31
Evacuation map
Network DisconnectedNetwork Connected
ENTER 2015 Research Track Slide Number 32
Safety ensuring by mail
No Problem
Injured
Illness
Stray
Lost Matter
Late
Other
Mr. A lost
his way in
Ginkaku-ji
Send Mail
Title : Safety
Confirmation
Are you all right?
Tell me your status.
No
Problem
In Trouble
Safety Confirm
Status Report
In case of
trouble
ENTER 2015 Research Track Slide Number 33
Outline of service specification
Before trip Planning in advance
- Display the observing attractions on the map.
During trip
Normal
Situation
Monitoring the students position.
- Track real-time students’ position. (per 1 second)
- Send students’ position to the server. (per 30 seconds)
- Store students’ trajectories in the remote server. (Tokyo)
Visualization of student trajectories.
- Indicate students’ current position and moving history.
Disaster
Situaion
Visualization of evacuation areas.
- Display the evacuation areas near position of students/teachers.
- Keep map display in case of the wireless network disconnection.
Voice & mail communication
- Broadcast confirmation mail to students from teachers.
- VoIP call among the permitted users.
ENTER 2015 Research Track Slide Number 34
GPS data
Date&Time Latitude Longitude Accuracy Provider BatteryLevel
2013/12/13 8:16 34.66693997 135.4960799 0GPS 86
2013/12/13 8:16 34.66884434 135.4969159 37GPS 86
2013/12/13 8:16 34.66915648 135.4967177 52GPS 86
2013/12/13 8:16 34.66908064 135.4967123 57GPS 86
2013/12/13 8:16 34.66900876 135.4968523 47GPS 86
2013/12/13 8:16 34.6692168 135.4962441 48GPS 86
2013/12/13 8:16 34.668968 135.496233 52GPS 86
2013/12/13 8:16 34.66880131 135.4963071 52GPS 86
2013/12/13 8:16 34.66868704 135.4963497 49GPS 86
2013/12/13 8:16 34.66827182 135.4963885 47GPS 86
2013/12/13 8:16 34.66789418 135.4962628 35GPS 86
2013/12/13 8:16 34.66777089 135.4962316 32GPS 86
2013/12/13 8:17 34.66764224 135.4961837 26GPS 86
2013/12/13 8:17 34.66750577 135.4961392 21GPS 86
2013/12/13 8:17 34.66743723 135.49611 21GPS 86
2013/12/13 8:17 34.66739099 135.496083 20GPS 86
2013/12/13 8:17 34.66731841 135.4960461 19GPS 86
2013/12/13 8:17 34.66729817 135.4960326 19GPS 86
2013/12/13 8:17 34.6673011 135.4960286 18GPS 86
2013/12/13 8:17 34.66731857 135.4960296 18GPS 86
2013/12/13 8:17 34.66732602 135.4960259 18GPS 86
2013/12/13 8:17 34.66732767 135.4960237 15GPS 86
2013/12/13 8:17 34.66732837 135.4960193 14GPS 86
2013/12/13 8:17 34.66732553 135.496015 14GPS 86
2013/12/13 8:17 34.66732211 135.4960096 14GPS 86
2013/12/13 8:17 34.66732243 135.4960029 14GPS 86
2013/12/13 8:17 34.66732497 135.4959998 14GPS 86
2013/12/13 8:17 34.66733029 135.4959973 14GPS 86
2013/12/13 8:17 34.66734217 135.4959961 14GPS 86
ENTER 2015 Research Track Slide Number 35
Comparison Experiment
without Environmental Constraints
Bus Taxi Train Walk
Total
count
Bus
44.3%
(8,353)
41.8%
(3,372)
1.7%
(83)
0.2%
(212) (22,831)
Taxi
1.6%
(303)
2.5%
(198)
19.3%
(964)
0.3%
(333)
(6,905)
Train
0%
(0)
0.3%
(21)
49.2%
(2,463)
0.5%
(598)
(5,540)
Walk
54.0%
(10,181)
55.4%
(4,469)
29.9%
(1,497)
99.1%
(127,083)
(124,854)
Total
count
100.0%
(18,837)
100.0%
(8,060)
100.0%
(5,007)
100.0%
(128,226)
86.2%
(160,130)
Estimate
Actual
*) Bus route constraints are used for bus / taxi estimation.
*) The universe is biased to “walk” mode.

More Related Content

What's hot

Traffic & Transportation surveys
Traffic & Transportation surveysTraffic & Transportation surveys
Traffic & Transportation surveysDhwani Shah
 
O and d study
O and d studyO and d study
O and d studymeetmksvs
 
Comparative study of emission pollutants between BIM and VSP methods.
Comparative study of emission pollutants between BIM and VSP methods.Comparative study of emission pollutants between BIM and VSP methods.
Comparative study of emission pollutants between BIM and VSP methods.AdithCR1
 
Aimsun saturadion flow rate calibration
Aimsun saturadion flow rate calibrationAimsun saturadion flow rate calibration
Aimsun saturadion flow rate calibrationJumpingJaq
 
myBas driving cycle for Kuala Terengganu city
myBas driving cycle for Kuala Terengganu city myBas driving cycle for Kuala Terengganu city
myBas driving cycle for Kuala Terengganu city IJECEIAES
 
Urban Road Congestion Management - Capacity Investments and Pricing Policies
Urban Road Congestion Management - Capacity Investments and Pricing PoliciesUrban Road Congestion Management - Capacity Investments and Pricing Policies
Urban Road Congestion Management - Capacity Investments and Pricing PoliciesBRTCoE
 
Spot speed studies and speed delay time survey
Spot speed studies and speed delay time surveySpot speed studies and speed delay time survey
Spot speed studies and speed delay time surveySai Santosh Yakkali
 
Presentation TRB BRT China 20170111
Presentation TRB BRT China 20170111Presentation TRB BRT China 20170111
Presentation TRB BRT China 20170111Pablo Guarda
 
Chapter 3&amp;4
Chapter 3&amp;4Chapter 3&amp;4
Chapter 3&amp;4EWIT
 
Hk icth2016 14th_june2016_htw_website version
Hk icth2016 14th_june2016_htw_website versionHk icth2016 14th_june2016_htw_website version
Hk icth2016 14th_june2016_htw_website versionHaneen Khreis
 
Origin and destination survey
Origin and destination surveyOrigin and destination survey
Origin and destination surveyraj balar
 
Origin & destination survey
Origin & destination surveyOrigin & destination survey
Origin & destination surveyAkash Pandey
 
(Slides) A Personal Navigation System with a Schedule Planning Facility Based...
(Slides) A Personal Navigation System with a Schedule Planning Facility Based...(Slides) A Personal Navigation System with a Schedule Planning Facility Based...
(Slides) A Personal Navigation System with a Schedule Planning Facility Based...Naoki Shibata
 
Traffic & transportation – ii
Traffic & transportation – iiTraffic & transportation – ii
Traffic & transportation – iiprahlad reddy
 
Student Pleanáil Submission by Gary Desmond (DIT) (3)
Student Pleanáil Submission by Gary Desmond (DIT) (3)Student Pleanáil Submission by Gary Desmond (DIT) (3)
Student Pleanáil Submission by Gary Desmond (DIT) (3)Gary Desmond
 
What You See is What You Get: In-use Measurement of Vehicle Activity, Emissio...
What You See is What You Get: In-use Measurement of Vehicle Activity, Emissio...What You See is What You Get: In-use Measurement of Vehicle Activity, Emissio...
What You See is What You Get: In-use Measurement of Vehicle Activity, Emissio...WRI Ross Center for Sustainable Cities
 
Transportation planning 1
Transportation planning 1Transportation planning 1
Transportation planning 1EngrABRahimoon
 
Freeway LOS (Transportation Engineering)
Freeway LOS (Transportation Engineering)Freeway LOS (Transportation Engineering)
Freeway LOS (Transportation Engineering)Hossam Shafiq I
 

What's hot (20)

Traffic & Transportation surveys
Traffic & Transportation surveysTraffic & Transportation surveys
Traffic & Transportation surveys
 
O and d study
O and d studyO and d study
O and d study
 
Comparative study of emission pollutants between BIM and VSP methods.
Comparative study of emission pollutants between BIM and VSP methods.Comparative study of emission pollutants between BIM and VSP methods.
Comparative study of emission pollutants between BIM and VSP methods.
 
Aimsun saturadion flow rate calibration
Aimsun saturadion flow rate calibrationAimsun saturadion flow rate calibration
Aimsun saturadion flow rate calibration
 
myBas driving cycle for Kuala Terengganu city
myBas driving cycle for Kuala Terengganu city myBas driving cycle for Kuala Terengganu city
myBas driving cycle for Kuala Terengganu city
 
Urban Road Congestion Management - Capacity Investments and Pricing Policies
Urban Road Congestion Management - Capacity Investments and Pricing PoliciesUrban Road Congestion Management - Capacity Investments and Pricing Policies
Urban Road Congestion Management - Capacity Investments and Pricing Policies
 
Spot speed studies and speed delay time survey
Spot speed studies and speed delay time surveySpot speed studies and speed delay time survey
Spot speed studies and speed delay time survey
 
He final ppt
He final pptHe final ppt
He final ppt
 
Cars cars everywhere
Cars cars everywhereCars cars everywhere
Cars cars everywhere
 
Presentation TRB BRT China 20170111
Presentation TRB BRT China 20170111Presentation TRB BRT China 20170111
Presentation TRB BRT China 20170111
 
Chapter 3&amp;4
Chapter 3&amp;4Chapter 3&amp;4
Chapter 3&amp;4
 
Hk icth2016 14th_june2016_htw_website version
Hk icth2016 14th_june2016_htw_website versionHk icth2016 14th_june2016_htw_website version
Hk icth2016 14th_june2016_htw_website version
 
Origin and destination survey
Origin and destination surveyOrigin and destination survey
Origin and destination survey
 
Origin & destination survey
Origin & destination surveyOrigin & destination survey
Origin & destination survey
 
(Slides) A Personal Navigation System with a Schedule Planning Facility Based...
(Slides) A Personal Navigation System with a Schedule Planning Facility Based...(Slides) A Personal Navigation System with a Schedule Planning Facility Based...
(Slides) A Personal Navigation System with a Schedule Planning Facility Based...
 
Traffic & transportation – ii
Traffic & transportation – iiTraffic & transportation – ii
Traffic & transportation – ii
 
Student Pleanáil Submission by Gary Desmond (DIT) (3)
Student Pleanáil Submission by Gary Desmond (DIT) (3)Student Pleanáil Submission by Gary Desmond (DIT) (3)
Student Pleanáil Submission by Gary Desmond (DIT) (3)
 
What You See is What You Get: In-use Measurement of Vehicle Activity, Emissio...
What You See is What You Get: In-use Measurement of Vehicle Activity, Emissio...What You See is What You Get: In-use Measurement of Vehicle Activity, Emissio...
What You See is What You Get: In-use Measurement of Vehicle Activity, Emissio...
 
Transportation planning 1
Transportation planning 1Transportation planning 1
Transportation planning 1
 
Freeway LOS (Transportation Engineering)
Freeway LOS (Transportation Engineering)Freeway LOS (Transportation Engineering)
Freeway LOS (Transportation Engineering)
 

Viewers also liked

Viewers also liked (20)

Applying IT on Promoting Egyptian Tourist Destination
Applying IT on Promoting Egyptian Tourist DestinationApplying IT on Promoting Egyptian Tourist Destination
Applying IT on Promoting Egyptian Tourist Destination
 
Online Learning and MOOCs: A Framework Proposal
Online Learning and MOOCs: A Framework ProposalOnline Learning and MOOCs: A Framework Proposal
Online Learning and MOOCs: A Framework Proposal
 
Technology as a Catalyst of Change: Enablers and Barriers of the Tourist Expe...
Technology as a Catalyst of Change: Enablers and Barriers of the Tourist Expe...Technology as a Catalyst of Change: Enablers and Barriers of the Tourist Expe...
Technology as a Catalyst of Change: Enablers and Barriers of the Tourist Expe...
 
The impact of attribute preferences on adoption timing of hotel distribution ...
The impact of attribute preferences on adoption timing of hotel distribution ...The impact of attribute preferences on adoption timing of hotel distribution ...
The impact of attribute preferences on adoption timing of hotel distribution ...
 
Strategic e-tourism Options for Destinations. Austrian Case
Strategic e-tourism Options for Destinations. Austrian CaseStrategic e-tourism Options for Destinations. Austrian Case
Strategic e-tourism Options for Destinations. Austrian Case
 
Crack the code: the phenomenon of Influence
Crack the code: the phenomenon of InfluenceCrack the code: the phenomenon of Influence
Crack the code: the phenomenon of Influence
 
Offline vs. online intermediation: a study of booking behaviour of tourists t...
Offline vs. online intermediation: a study of booking behaviour of tourists t...Offline vs. online intermediation: a study of booking behaviour of tourists t...
Offline vs. online intermediation: a study of booking behaviour of tourists t...
 
An Exploratory Study on Drivers and Deterrents of Collaborative Consumption ...
An Exploratory Study on Drivers and  Deterrents of Collaborative Consumption ...An Exploratory Study on Drivers and  Deterrents of Collaborative Consumption ...
An Exploratory Study on Drivers and Deterrents of Collaborative Consumption ...
 
The Evolution of eTourism Research A Case of ENTER Conference
The Evolution of eTourism Research A Case of ENTER ConferenceThe Evolution of eTourism Research A Case of ENTER Conference
The Evolution of eTourism Research A Case of ENTER Conference
 
A practical approach to big data in tourism: a low cost Raspberry Pi cluster
A practical approach to big data in tourism: a low cost Raspberry Pi clusterA practical approach to big data in tourism: a low cost Raspberry Pi cluster
A practical approach to big data in tourism: a low cost Raspberry Pi cluster
 
Brennpunkt2015 Lassnig
Brennpunkt2015 LassnigBrennpunkt2015 Lassnig
Brennpunkt2015 Lassnig
 
Online Tourism Review: Three Phases for Successful Destination Relationships
Online Tourism Review: Three Phases for Successful Destination RelationshipsOnline Tourism Review: Three Phases for Successful Destination Relationships
Online Tourism Review: Three Phases for Successful Destination Relationships
 
Promoting Religious Tourism as a new niche Tourism in Lebanon using ICT
Promoting Religious Tourism as a new niche Tourism in Lebanon using ICTPromoting Religious Tourism as a new niche Tourism in Lebanon using ICT
Promoting Religious Tourism as a new niche Tourism in Lebanon using ICT
 
An Investigation of Features and Functions of Smartphone Applications for Hot...
An Investigation of Features and Functions of Smartphone Applications for Hot...An Investigation of Features and Functions of Smartphone Applications for Hot...
An Investigation of Features and Functions of Smartphone Applications for Hot...
 
An Application of Apriori Algorithm Association Rules Mining to Profiling the...
An Application of Apriori Algorithm Association Rules Mining to Profiling the...An Application of Apriori Algorithm Association Rules Mining to Profiling the...
An Application of Apriori Algorithm Association Rules Mining to Profiling the...
 
Tourists and Municipal Wi-Fi Networks. The case of Lugano (CH
Tourists and Municipal Wi-Fi Networks. The case of Lugano (CHTourists and Municipal Wi-Fi Networks. The case of Lugano (CH
Tourists and Municipal Wi-Fi Networks. The case of Lugano (CH
 
Behaviour of virtual visitor based on eShop and DMO websites: A comparative s...
Behaviour of virtual visitor based on eShop and DMO websites: A comparative s...Behaviour of virtual visitor based on eShop and DMO websites: A comparative s...
Behaviour of virtual visitor based on eShop and DMO websites: A comparative s...
 
The new Ticino digital experience
The new Ticino digital experienceThe new Ticino digital experience
The new Ticino digital experience
 
Feefo
FeefoFeefo
Feefo
 
What Types of Hotels Make Their Guests (Un)Happy? Text Analytics of Customer ...
What Types of Hotels Make Their Guests (Un)Happy? Text Analytics of Customer ...What Types of Hotels Make Their Guests (Un)Happy? Text Analytics of Customer ...
What Types of Hotels Make Their Guests (Un)Happy? Text Analytics of Customer ...
 

Similar to Annotating Transportation Modes from Tourist GPS Trajectories

Updated Traffic Analysis Tools for Complete Streets
Updated Traffic Analysis Tools for Complete StreetsUpdated Traffic Analysis Tools for Complete Streets
Updated Traffic Analysis Tools for Complete StreetsWSP
 
INTELLIGENT TRANSPORTATION SYSTEM
INTELLIGENT TRANSPORTATION SYSTEMINTELLIGENT TRANSPORTATION SYSTEM
INTELLIGENT TRANSPORTATION SYSTEMAmar Patel
 
Transport solution for daily commuters from moratuwa area to colombo
Transport solution for daily commuters from moratuwa area to colombo Transport solution for daily commuters from moratuwa area to colombo
Transport solution for daily commuters from moratuwa area to colombo Praneeth Prabodha Dissanayaka, MILT
 
Kp rail visionevent_18nov2015
Kp rail visionevent_18nov2015Kp rail visionevent_18nov2015
Kp rail visionevent_18nov2015Barry Scott
 
The Swiss national stated preference study on transport behavior 2015
The Swiss national stated preference study on transport behavior 2015The Swiss national stated preference study on transport behavior 2015
The Swiss national stated preference study on transport behavior 2015Antonin Danalet
 
Webinar: Planning, design, implementation and operation of the Yichang BRT co...
Webinar: Planning, design, implementation and operation of the Yichang BRT co...Webinar: Planning, design, implementation and operation of the Yichang BRT co...
Webinar: Planning, design, implementation and operation of the Yichang BRT co...BRTCoE
 
Advance road transportation system ppt
Advance road transportation system pptAdvance road transportation system ppt
Advance road transportation system pptAbdul Aziz
 
[Transportation] 2. kulrathna(colombo)
[Transportation] 2. kulrathna(colombo)[Transportation] 2. kulrathna(colombo)
[Transportation] 2. kulrathna(colombo)shrdcinfo
 
Urban transport (MODAL SHIFT ANALYSIS)
Urban transport (MODAL SHIFT ANALYSIS)Urban transport (MODAL SHIFT ANALYSIS)
Urban transport (MODAL SHIFT ANALYSIS)syafiqahbahrin
 
Traffic Engineering And Drainage
Traffic Engineering And DrainageTraffic Engineering And Drainage
Traffic Engineering And DrainagePrashant Ranjan
 
Speed study pradipta banik 1204012
Speed study pradipta banik 1204012Speed study pradipta banik 1204012
Speed study pradipta banik 1204012Pradipta Banik
 
5d6a0c3d-4f6c-4e3f-b6ad-59fa23676177-160229074036
5d6a0c3d-4f6c-4e3f-b6ad-59fa23676177-1602290740365d6a0c3d-4f6c-4e3f-b6ad-59fa23676177-160229074036
5d6a0c3d-4f6c-4e3f-b6ad-59fa23676177-160229074036123456yy
 
TRAFFIC AND PEDESTRIAN MANAGEMENT FOR YESHWANTHPUR
TRAFFIC AND PEDESTRIAN MANAGEMENT FOR YESHWANTHPURTRAFFIC AND PEDESTRIAN MANAGEMENT FOR YESHWANTHPUR
TRAFFIC AND PEDESTRIAN MANAGEMENT FOR YESHWANTHPURPramod Shivanand
 
4. traffic engineering
4. traffic engineering4. traffic engineering
4. traffic engineeringholegajendra
 
Webinar: Land Use-Transport Interactions: Evidence from and Implications for ...
Webinar: Land Use-Transport Interactions: Evidence from and Implications for ...Webinar: Land Use-Transport Interactions: Evidence from and Implications for ...
Webinar: Land Use-Transport Interactions: Evidence from and Implications for ...BRTCoE
 
Analysis of the Pedestrian System In Jayapura City (A Case Study of Pedestria...
Analysis of the Pedestrian System In Jayapura City (A Case Study of Pedestria...Analysis of the Pedestrian System In Jayapura City (A Case Study of Pedestria...
Analysis of the Pedestrian System In Jayapura City (A Case Study of Pedestria...ijceronline
 
LO5: Simulation of transit signal priority strategies for brt operations
LO5: Simulation of transit signal priority strategies for brt operationsLO5: Simulation of transit signal priority strategies for brt operations
LO5: Simulation of transit signal priority strategies for brt operationsBRTCoE
 

Similar to Annotating Transportation Modes from Tourist GPS Trajectories (20)

Updated Traffic Analysis Tools for Complete Streets
Updated Traffic Analysis Tools for Complete StreetsUpdated Traffic Analysis Tools for Complete Streets
Updated Traffic Analysis Tools for Complete Streets
 
INTELLIGENT TRANSPORTATION SYSTEM
INTELLIGENT TRANSPORTATION SYSTEMINTELLIGENT TRANSPORTATION SYSTEM
INTELLIGENT TRANSPORTATION SYSTEM
 
Transport solution for daily commuters from moratuwa area to colombo
Transport solution for daily commuters from moratuwa area to colombo Transport solution for daily commuters from moratuwa area to colombo
Transport solution for daily commuters from moratuwa area to colombo
 
Kp rail visionevent_18nov2015
Kp rail visionevent_18nov2015Kp rail visionevent_18nov2015
Kp rail visionevent_18nov2015
 
The Swiss national stated preference study on transport behavior 2015
The Swiss national stated preference study on transport behavior 2015The Swiss national stated preference study on transport behavior 2015
The Swiss national stated preference study on transport behavior 2015
 
Webinar: Planning, design, implementation and operation of the Yichang BRT co...
Webinar: Planning, design, implementation and operation of the Yichang BRT co...Webinar: Planning, design, implementation and operation of the Yichang BRT co...
Webinar: Planning, design, implementation and operation of the Yichang BRT co...
 
Advance road transportation system ppt
Advance road transportation system pptAdvance road transportation system ppt
Advance road transportation system ppt
 
[Transportation] 2. kulrathna(colombo)
[Transportation] 2. kulrathna(colombo)[Transportation] 2. kulrathna(colombo)
[Transportation] 2. kulrathna(colombo)
 
Urban transport (MODAL SHIFT ANALYSIS)
Urban transport (MODAL SHIFT ANALYSIS)Urban transport (MODAL SHIFT ANALYSIS)
Urban transport (MODAL SHIFT ANALYSIS)
 
Traffic Engineering And Drainage
Traffic Engineering And DrainageTraffic Engineering And Drainage
Traffic Engineering And Drainage
 
New Tools for Estimating Walking and Bicycling Demand
New Tools for Estimating Walking and Bicycling DemandNew Tools for Estimating Walking and Bicycling Demand
New Tools for Estimating Walking and Bicycling Demand
 
Speed study pradipta banik 1204012
Speed study pradipta banik 1204012Speed study pradipta banik 1204012
Speed study pradipta banik 1204012
 
5d6a0c3d-4f6c-4e3f-b6ad-59fa23676177-160229074036
5d6a0c3d-4f6c-4e3f-b6ad-59fa23676177-1602290740365d6a0c3d-4f6c-4e3f-b6ad-59fa23676177-160229074036
5d6a0c3d-4f6c-4e3f-b6ad-59fa23676177-160229074036
 
TRAFFIC AND PEDESTRIAN MANAGEMENT FOR YESHWANTHPUR
TRAFFIC AND PEDESTRIAN MANAGEMENT FOR YESHWANTHPURTRAFFIC AND PEDESTRIAN MANAGEMENT FOR YESHWANTHPUR
TRAFFIC AND PEDESTRIAN MANAGEMENT FOR YESHWANTHPUR
 
dr. cal
dr. caldr. cal
dr. cal
 
4. traffic engineering
4. traffic engineering4. traffic engineering
4. traffic engineering
 
June 15, 2021 BPAC Virtual Workshop
June 15, 2021 BPAC Virtual Workshop June 15, 2021 BPAC Virtual Workshop
June 15, 2021 BPAC Virtual Workshop
 
Webinar: Land Use-Transport Interactions: Evidence from and Implications for ...
Webinar: Land Use-Transport Interactions: Evidence from and Implications for ...Webinar: Land Use-Transport Interactions: Evidence from and Implications for ...
Webinar: Land Use-Transport Interactions: Evidence from and Implications for ...
 
Analysis of the Pedestrian System In Jayapura City (A Case Study of Pedestria...
Analysis of the Pedestrian System In Jayapura City (A Case Study of Pedestria...Analysis of the Pedestrian System In Jayapura City (A Case Study of Pedestria...
Analysis of the Pedestrian System In Jayapura City (A Case Study of Pedestria...
 
LO5: Simulation of transit signal priority strategies for brt operations
LO5: Simulation of transit signal priority strategies for brt operationsLO5: Simulation of transit signal priority strategies for brt operations
LO5: Simulation of transit signal priority strategies for brt operations
 

Recently uploaded

Interactive Powerpoint_How to Master effective communication
Interactive Powerpoint_How to Master effective communicationInteractive Powerpoint_How to Master effective communication
Interactive Powerpoint_How to Master effective communicationnomboosow
 
Mastering the Unannounced Regulatory Inspection
Mastering the Unannounced Regulatory InspectionMastering the Unannounced Regulatory Inspection
Mastering the Unannounced Regulatory InspectionSafetyChain Software
 
The basics of sentences session 2pptx copy.pptx
The basics of sentences session 2pptx copy.pptxThe basics of sentences session 2pptx copy.pptx
The basics of sentences session 2pptx copy.pptxheathfieldcps1
 
Paris 2024 Olympic Geographies - an activity
Paris 2024 Olympic Geographies - an activityParis 2024 Olympic Geographies - an activity
Paris 2024 Olympic Geographies - an activityGeoBlogs
 
Sanyam Choudhary Chemistry practical.pdf
Sanyam Choudhary Chemistry practical.pdfSanyam Choudhary Chemistry practical.pdf
Sanyam Choudhary Chemistry practical.pdfsanyamsingh5019
 
Web & Social Media Analytics Previous Year Question Paper.pdf
Web & Social Media Analytics Previous Year Question Paper.pdfWeb & Social Media Analytics Previous Year Question Paper.pdf
Web & Social Media Analytics Previous Year Question Paper.pdfJayanti Pande
 
Presentation by Andreas Schleicher Tackling the School Absenteeism Crisis 30 ...
Presentation by Andreas Schleicher Tackling the School Absenteeism Crisis 30 ...Presentation by Andreas Schleicher Tackling the School Absenteeism Crisis 30 ...
Presentation by Andreas Schleicher Tackling the School Absenteeism Crisis 30 ...EduSkills OECD
 
Advanced Views - Calendar View in Odoo 17
Advanced Views - Calendar View in Odoo 17Advanced Views - Calendar View in Odoo 17
Advanced Views - Calendar View in Odoo 17Celine George
 
Introduction to ArtificiaI Intelligence in Higher Education
Introduction to ArtificiaI Intelligence in Higher EducationIntroduction to ArtificiaI Intelligence in Higher Education
Introduction to ArtificiaI Intelligence in Higher Educationpboyjonauth
 
Industrial Policy - 1948, 1956, 1973, 1977, 1980, 1991
Industrial Policy - 1948, 1956, 1973, 1977, 1980, 1991Industrial Policy - 1948, 1956, 1973, 1977, 1980, 1991
Industrial Policy - 1948, 1956, 1973, 1977, 1980, 1991RKavithamani
 
microwave assisted reaction. General introduction
microwave assisted reaction. General introductionmicrowave assisted reaction. General introduction
microwave assisted reaction. General introductionMaksud Ahmed
 
Software Engineering Methodologies (overview)
Software Engineering Methodologies (overview)Software Engineering Methodologies (overview)
Software Engineering Methodologies (overview)eniolaolutunde
 
Activity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdfActivity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdfciinovamais
 
mini mental status format.docx
mini    mental       status     format.docxmini    mental       status     format.docx
mini mental status format.docxPoojaSen20
 
SOCIAL AND HISTORICAL CONTEXT - LFTVD.pptx
SOCIAL AND HISTORICAL CONTEXT - LFTVD.pptxSOCIAL AND HISTORICAL CONTEXT - LFTVD.pptx
SOCIAL AND HISTORICAL CONTEXT - LFTVD.pptxiammrhaywood
 
POINT- BIOCHEMISTRY SEM 2 ENZYMES UNIT 5.pptx
POINT- BIOCHEMISTRY SEM 2 ENZYMES UNIT 5.pptxPOINT- BIOCHEMISTRY SEM 2 ENZYMES UNIT 5.pptx
POINT- BIOCHEMISTRY SEM 2 ENZYMES UNIT 5.pptxSayali Powar
 
1029 - Danh muc Sach Giao Khoa 10 . pdf
1029 -  Danh muc Sach Giao Khoa 10 . pdf1029 -  Danh muc Sach Giao Khoa 10 . pdf
1029 - Danh muc Sach Giao Khoa 10 . pdfQucHHunhnh
 
Privatization and Disinvestment - Meaning, Objectives, Advantages and Disadva...
Privatization and Disinvestment - Meaning, Objectives, Advantages and Disadva...Privatization and Disinvestment - Meaning, Objectives, Advantages and Disadva...
Privatization and Disinvestment - Meaning, Objectives, Advantages and Disadva...RKavithamani
 
Nutritional Needs Presentation - HLTH 104
Nutritional Needs Presentation - HLTH 104Nutritional Needs Presentation - HLTH 104
Nutritional Needs Presentation - HLTH 104misteraugie
 

Recently uploaded (20)

Interactive Powerpoint_How to Master effective communication
Interactive Powerpoint_How to Master effective communicationInteractive Powerpoint_How to Master effective communication
Interactive Powerpoint_How to Master effective communication
 
Mastering the Unannounced Regulatory Inspection
Mastering the Unannounced Regulatory InspectionMastering the Unannounced Regulatory Inspection
Mastering the Unannounced Regulatory Inspection
 
The basics of sentences session 2pptx copy.pptx
The basics of sentences session 2pptx copy.pptxThe basics of sentences session 2pptx copy.pptx
The basics of sentences session 2pptx copy.pptx
 
Paris 2024 Olympic Geographies - an activity
Paris 2024 Olympic Geographies - an activityParis 2024 Olympic Geographies - an activity
Paris 2024 Olympic Geographies - an activity
 
Sanyam Choudhary Chemistry practical.pdf
Sanyam Choudhary Chemistry practical.pdfSanyam Choudhary Chemistry practical.pdf
Sanyam Choudhary Chemistry practical.pdf
 
Web & Social Media Analytics Previous Year Question Paper.pdf
Web & Social Media Analytics Previous Year Question Paper.pdfWeb & Social Media Analytics Previous Year Question Paper.pdf
Web & Social Media Analytics Previous Year Question Paper.pdf
 
Presentation by Andreas Schleicher Tackling the School Absenteeism Crisis 30 ...
Presentation by Andreas Schleicher Tackling the School Absenteeism Crisis 30 ...Presentation by Andreas Schleicher Tackling the School Absenteeism Crisis 30 ...
Presentation by Andreas Schleicher Tackling the School Absenteeism Crisis 30 ...
 
Staff of Color (SOC) Retention Efforts DDSD
Staff of Color (SOC) Retention Efforts DDSDStaff of Color (SOC) Retention Efforts DDSD
Staff of Color (SOC) Retention Efforts DDSD
 
Advanced Views - Calendar View in Odoo 17
Advanced Views - Calendar View in Odoo 17Advanced Views - Calendar View in Odoo 17
Advanced Views - Calendar View in Odoo 17
 
Introduction to ArtificiaI Intelligence in Higher Education
Introduction to ArtificiaI Intelligence in Higher EducationIntroduction to ArtificiaI Intelligence in Higher Education
Introduction to ArtificiaI Intelligence in Higher Education
 
Industrial Policy - 1948, 1956, 1973, 1977, 1980, 1991
Industrial Policy - 1948, 1956, 1973, 1977, 1980, 1991Industrial Policy - 1948, 1956, 1973, 1977, 1980, 1991
Industrial Policy - 1948, 1956, 1973, 1977, 1980, 1991
 
microwave assisted reaction. General introduction
microwave assisted reaction. General introductionmicrowave assisted reaction. General introduction
microwave assisted reaction. General introduction
 
Software Engineering Methodologies (overview)
Software Engineering Methodologies (overview)Software Engineering Methodologies (overview)
Software Engineering Methodologies (overview)
 
Activity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdfActivity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdf
 
mini mental status format.docx
mini    mental       status     format.docxmini    mental       status     format.docx
mini mental status format.docx
 
SOCIAL AND HISTORICAL CONTEXT - LFTVD.pptx
SOCIAL AND HISTORICAL CONTEXT - LFTVD.pptxSOCIAL AND HISTORICAL CONTEXT - LFTVD.pptx
SOCIAL AND HISTORICAL CONTEXT - LFTVD.pptx
 
POINT- BIOCHEMISTRY SEM 2 ENZYMES UNIT 5.pptx
POINT- BIOCHEMISTRY SEM 2 ENZYMES UNIT 5.pptxPOINT- BIOCHEMISTRY SEM 2 ENZYMES UNIT 5.pptx
POINT- BIOCHEMISTRY SEM 2 ENZYMES UNIT 5.pptx
 
1029 - Danh muc Sach Giao Khoa 10 . pdf
1029 -  Danh muc Sach Giao Khoa 10 . pdf1029 -  Danh muc Sach Giao Khoa 10 . pdf
1029 - Danh muc Sach Giao Khoa 10 . pdf
 
Privatization and Disinvestment - Meaning, Objectives, Advantages and Disadva...
Privatization and Disinvestment - Meaning, Objectives, Advantages and Disadva...Privatization and Disinvestment - Meaning, Objectives, Advantages and Disadva...
Privatization and Disinvestment - Meaning, Objectives, Advantages and Disadva...
 
Nutritional Needs Presentation - HLTH 104
Nutritional Needs Presentation - HLTH 104Nutritional Needs Presentation - HLTH 104
Nutritional Needs Presentation - HLTH 104
 

Annotating Transportation Modes from Tourist GPS Trajectories

  • 1. ENTER 2015 Research Track Slide Number 1 Transportation Mode Annotation of Tourist GPS Trajectories under Environmental Constraints Hidekazu Kasahara, Mikihiko Mori, Masayuki Mukunoki, and Michihiko Minoh Kyoto University, Japan Hidekazu.kasahara@mm.media.kyoto-u.ac.jp http://www.mm.media.kyoto-u.ac.jp/en/members/hidekazu- kasahara
  • 2. ENTER 2015 Research Track Slide Number 2 Agenda • Motivation • Purpose • Previous Work • Problems • Proposed Method • Experiment & Evaluation • Discussion • Future Work & Conclusion
  • 3. ENTER 2015 Research Track Slide Number 3 Motivation • DMOs want to know the tourists’ activities. – Policy making / destination marketing strategy. • Optimum deployment of transportation. • Outdoor ads / direction boards. Spot B Spot A Bus 1,000 tourists Taxi 10 tourists Ads on bus is more effective than those on taxi in this route. Accurate transportation usage statistics is necessary for DMOs for the decision making. The traditional stats is costly.
  • 4. ENTER 2015 Research Track Slide Number 4 Purpose of Study • Annotating transportation mode from tourists’ GPS trajectory (x, y, t). • Modeling the tourists activity as a chain of transportation modes. • Transportation usage statistics is calculated from these chains of transportation modes. • In addition, tourists’ personal preferences can be explored from the chains. Walk 60 min. Bus 20 min.Bus 20 min. WalkWalk 5 min. Train 20 min.
  • 5. ENTER 2015 Research Track Slide Number 5 Scope of Study • Tourists move by foot and public transportation. – Train, route bus and taxi as public transportation. – Tour bus, bicycle, motor-bike and hire car are out of scope. • Destinations that include numerous locations of interest, such as Milano. • Only GPS. – Acceleration meter is out of scope.
  • 6. ENTER 2015 Research Track Slide Number 6 Previous Work • Approaches using tourist moving physical features. – Speed: Decision tree (Aoki 2008, Zheng 2010). – Baysean network (Stenneth 2011). – Hierarchical CRF (Liao 2007). – SVM (Bolbol 2012). – Acceleration: Adaptive boosting (Hemminki 2013). – Regression analysis (Furukawa 2014). • Approach using environmental factors. – Assumption on typical tourist travel patterns (Stopher 2008). – Bus stops and road network (Liao 2007). – Land utilization situation (Yan 2013).
  • 7. ENTER 2015 Research Track Slide Number 7 Transportation Mode Estimation using Tourist Speed Yellow points shows deceleration at bus stops and crossings. Station Temple Castle Train Golden Pagoda Yellow: Under 1km/hour Black : 1-60 km/hour Red : Over 60 km/hour These decelerations are estimated as walk / retain modes using by speed. GPS Trajectory 1km
  • 8. ENTER 2015 Research Track Slide Number 8 Annotation using Environmental Constraints bus bus train train train bus p1 p2 p3 p4 p5 p6 Estimate using Tourist Features (Speed) • In physical space, there are some environmental constraints restrict transportation usage. – Environmental constraints.
  • 9. ENTER 2015 Research Track Slide Number 9 Annotation using Environmental Constraints Environmental Constraints : All 6 Points Are on a Bus Route Not on a railway. Estimate using Tourist Features (Speed) • In physical space, there are some environmental constraints restrict transportation usage. – Environmental constraints. bus bus train train train bus p1 p2 p3 p4 p5 p6
  • 10. ENTER 2015 Research Track Slide Number 10 Annotation using Environmental Constraints Environmental Constraints : All 6 Points Are on a Bus Route Not on a railway. Estimate using Tourist Features (Speed) Inconsistency • In physical space, there are some environmental constraints restrict transportation usage. – Environmental constraints. • In the case when the estimation with tourist moving features is failed, there is regional inconsistency with environmental constraints. bus bus train train train bus p1 p2 p3 p4 p5 p6
  • 11. ENTER 2015 Research Track Slide Number 11 Problem • The system failed to estimate car and train as retain or walk modes when they decelerate or stop in case the system use only tourist features (speed). • Examples of deceleration: – Route bus decelerates at all bus stop along the route. – Train decelerates at stations on the railways.
  • 12. ENTER 2015 Research Track Slide Number 12 Proposed Method • Using the environmental constraints, decrease the inconsistency between the estimate using tourist features (speed) and environmental constraints. • Annotate the transportation modes with the least inconsistency in total. • We evaluate the proposed methods compared with human labeling.
  • 13. ENTER 2015 Research Track Slide Number 13 Proposed Method Output : GPS Trajectory Consists of (x, y, time, speed) • Environmental constraints • Speed Output: Fragmented Segments Output: Merged Segments with the least inconsistency in total • Interleave • Continuity • Filtering • Speed calculation Technique ‫ݏ‬ଵ ‫ݏ‬ଶ ‫ݏ‬ଷ ‫ݏ‬ସ ‫ݏ‬ହ ݉ଵ ݉ଶ ݉ଵ ݉ଶ ݉ଵ ‫ݏ‬ଵ ܵଶ ᇱ ‫ݏ‬ହ ݉ଵ ݉ଶ ݉ଵ Preprocess Stage 1 Stage 2
  • 14. ENTER 2015 Research Track Slide Number 14 Preprocess & Stage #1 (x, y, time) : GPS raw data (x, y, time, speed) Filtering & speed calculation On Train Route On Bus Route On Motorway On Pedestrian area (x, y, time, speed, mode, segment) speed > sp_train Speed Environmental Constraints sp_walk < speed speed > sp_walk and and … Stop frequency
  • 15. ENTER 2015 Research Track Slide Number 15 #2 Stage: Segment Merger • Interleave Assumption: Tourists do not change “train to car” or “car to train” directly. • Continuity Assumption: Tourists typically do not change the transportation mode over a short period. Bus Taxi Walk Bus Never Directly Connected Too short to change
  • 16. ENTER 2015 Research Track Slide Number 16 Experiment • Dataset – 16 persons’ GPS trajectory data from students and teachers who travelled. • Destination : Kyoto • 160,130 points in total. • GPS measures location per 1 second. • From 8:00 to 18:00.
  • 17. ENTER 2015 Research Track Slide Number 17 Screening Value • sp_walk = 2 metres per second. – The screening value between walk and bus/taxi. • sp_train = 15 metres per second. – The screening value between bus/taxi and train. • ݀݁݊‫݁ݏ‬ = 10% – The screening value of stop frequency between bus and taxi. (Route bus stops more frequent than taxi.) • Interleave = 240 seconds – If the length of segment is under 240 seconds, the segment is merged to next or before segment
  • 18. ENTER 2015 Research Track Slide Number 18 Environmental Constraints • Data is supplied as shape files. Bus Route (Gov. data) Train Route (COTS) Motorway (COTS) Pedestrian Zone (Gov. data)
  • 19. ENTER 2015 Research Track Slide Number 19 Results of the Experiment Bus Taxi Train Walk Total count Bus 82.8% (15,601) 17.5% (1,413) 0.0% (0) 4.5% (5,817) (22,831) Taxi 6.1% (1,158) 60.1% (4,847) 10.6% (529) 0.3% (371) (6,905) Train 0.2% (30) 0.3% (21) 77.3% (3,869) 1.3% (1,620) (5,540) Walk 10.9% (2,048) 22.1% (1,779) 12.2% (609) 93.9% (120,418) (124,854) Total count 100.0% (18,837) 100.0% (8,060) 100.0% (5,007) 100.0% (128,226) 90.4% (160,130) Estimate Actual
  • 20. ENTER 2015 Research Track Slide Number 20 Comparison Bus Taxi Train Walk Total Proposed 82.8% 60.1% 77.3% 93.9% 90.4% SVM *1) 88.9 % 78.9% 67.1% 67.7% 75.2% Speed without environmental constraints *2)*3) 44.3% 2.5% 49.2% 99.1% 86.2% Method Mode *1) Result of 5-fold cross validation on the same data set *2) Bus route constraints are used for bus / taxi estimation. *3) The universe is biased to “walk” mode.
  • 21. ENTER 2015 Research Track Slide Number 21 Discussion • Automatic estimation – The screening values are given by human in the proposed method. • Generalization ability is low. • These values should be decided automatically. – Stochastic function. • Function can be estimated from training data. • Stochastic function varies in different environments 0% 20% 40% 60% 80% 100% Mode Probability Function Walk Tran Taxi Bus
  • 22. ENTER 2015 Research Track Slide Number 22 Discussion • Context among primitive components – Transportation modes consist of three primitive components; “stable speed move”, “acceleration (deceleration)” and “stop.” – We call context to the relationship between consecutive segments of each primitive component. – We should consider the context for estimation. – We have two assumptions of context in this study and give some screening values. However, these should be automatically decided by using training data.
  • 23. ENTER 2015 Research Track Slide Number 23 Discussion • Distinction of bus and taxi is difficult – Only using speed, this system cannot distinguish between taxi and bus accurately. • Other environmental constraints is necessary. – Other environmental constraints is necessary. • Other environmental constraints – Traffic confusion. – Bus stop location.
  • 24. ENTER 2015 Research Track Slide Number 24 Conclusion & Future Work • We proposed a new GPS semantic annotation method using environmental constraints based on two assumptions of tourist behaviour. • The results indicate the high accuracy, 90%. • Making model based on stochastic logic. – Improving generalization capability of the method. • By using other environmental constraints, we try to improve the performance in future.
  • 25. ENTER 2015 Research Track Slide Number 25 Appendix
  • 26. ENTER 2015 Research Track Slide Number 26 Transportation Mode Annotation using Tourist Features Annotation using tourist’s speed is valid for retain / walk modes. Golden Pagoda Blue : 10-60km/hour Yellow: Under 1km/hour Green : 1-10 km/hour In this area, system estimated retain / walk / bus modes. GPS Trajectory
  • 27. ENTER 2015 Research Track Slide Number 27 Transportation Mode Estimation from GPS Trajectory Walk 60 min. Bus 20 min. Chain of Transportation Modes Bus 20 min. Bus GPS Trajectory Walk Temple
  • 28. ENTER 2015 Research Track Slide Number 28 What’s school trip? School trip is one of the biggest group tours in Japan • School trip is important for DMOs and travel agents • 10% of all stayed tourists in Kyoto 2012 is school trip students • The number of students who participated school trip in 2012 is 3.4 million • Participation rate of students is high • 94.4% of junior high schools /75.5% of senior high schools Unit : million
  • 29. ENTER 2015 Research Track Slide Number 29 Escort-teachers Field HQ (HOTEL) School staff School(Home) School master Real-time monitoring Current position Smartphone (Group leader carries) Trajectory Tablet/PC - GPS & Wi-Fi positioning Overview of safety ensuring system ETSS (Educational Tour Support System) Student groups Information Sharing among All Related Persons Group Leader(Student. Trained before trip.) - Mail & Voice Safety ensuring No Navigation For education System Tablet
  • 30. ENTER 2015 Research Track Slide Number 30 Screenshot of student phone Schoolmaster Homeroom teacher Voice & MailDisaster MapNormal Map School NameNow 1 hr All EvacuationArea ObservingAttractions Current Position Evacuation Area MapNormal MapNormal Map
  • 31. ENTER 2015 Research Track Slide Number 31 Evacuation map Network DisconnectedNetwork Connected
  • 32. ENTER 2015 Research Track Slide Number 32 Safety ensuring by mail No Problem Injured Illness Stray Lost Matter Late Other Mr. A lost his way in Ginkaku-ji Send Mail Title : Safety Confirmation Are you all right? Tell me your status. No Problem In Trouble Safety Confirm Status Report In case of trouble
  • 33. ENTER 2015 Research Track Slide Number 33 Outline of service specification Before trip Planning in advance - Display the observing attractions on the map. During trip Normal Situation Monitoring the students position. - Track real-time students’ position. (per 1 second) - Send students’ position to the server. (per 30 seconds) - Store students’ trajectories in the remote server. (Tokyo) Visualization of student trajectories. - Indicate students’ current position and moving history. Disaster Situaion Visualization of evacuation areas. - Display the evacuation areas near position of students/teachers. - Keep map display in case of the wireless network disconnection. Voice & mail communication - Broadcast confirmation mail to students from teachers. - VoIP call among the permitted users.
  • 34. ENTER 2015 Research Track Slide Number 34 GPS data Date&Time Latitude Longitude Accuracy Provider BatteryLevel 2013/12/13 8:16 34.66693997 135.4960799 0GPS 86 2013/12/13 8:16 34.66884434 135.4969159 37GPS 86 2013/12/13 8:16 34.66915648 135.4967177 52GPS 86 2013/12/13 8:16 34.66908064 135.4967123 57GPS 86 2013/12/13 8:16 34.66900876 135.4968523 47GPS 86 2013/12/13 8:16 34.6692168 135.4962441 48GPS 86 2013/12/13 8:16 34.668968 135.496233 52GPS 86 2013/12/13 8:16 34.66880131 135.4963071 52GPS 86 2013/12/13 8:16 34.66868704 135.4963497 49GPS 86 2013/12/13 8:16 34.66827182 135.4963885 47GPS 86 2013/12/13 8:16 34.66789418 135.4962628 35GPS 86 2013/12/13 8:16 34.66777089 135.4962316 32GPS 86 2013/12/13 8:17 34.66764224 135.4961837 26GPS 86 2013/12/13 8:17 34.66750577 135.4961392 21GPS 86 2013/12/13 8:17 34.66743723 135.49611 21GPS 86 2013/12/13 8:17 34.66739099 135.496083 20GPS 86 2013/12/13 8:17 34.66731841 135.4960461 19GPS 86 2013/12/13 8:17 34.66729817 135.4960326 19GPS 86 2013/12/13 8:17 34.6673011 135.4960286 18GPS 86 2013/12/13 8:17 34.66731857 135.4960296 18GPS 86 2013/12/13 8:17 34.66732602 135.4960259 18GPS 86 2013/12/13 8:17 34.66732767 135.4960237 15GPS 86 2013/12/13 8:17 34.66732837 135.4960193 14GPS 86 2013/12/13 8:17 34.66732553 135.496015 14GPS 86 2013/12/13 8:17 34.66732211 135.4960096 14GPS 86 2013/12/13 8:17 34.66732243 135.4960029 14GPS 86 2013/12/13 8:17 34.66732497 135.4959998 14GPS 86 2013/12/13 8:17 34.66733029 135.4959973 14GPS 86 2013/12/13 8:17 34.66734217 135.4959961 14GPS 86
  • 35. ENTER 2015 Research Track Slide Number 35 Comparison Experiment without Environmental Constraints Bus Taxi Train Walk Total count Bus 44.3% (8,353) 41.8% (3,372) 1.7% (83) 0.2% (212) (22,831) Taxi 1.6% (303) 2.5% (198) 19.3% (964) 0.3% (333) (6,905) Train 0% (0) 0.3% (21) 49.2% (2,463) 0.5% (598) (5,540) Walk 54.0% (10,181) 55.4% (4,469) 29.9% (1,497) 99.1% (127,083) (124,854) Total count 100.0% (18,837) 100.0% (8,060) 100.0% (5,007) 100.0% (128,226) 86.2% (160,130) Estimate Actual *) Bus route constraints are used for bus / taxi estimation. *) The universe is biased to “walk” mode.