SlideShare ist ein Scribd-Unternehmen logo
1 von 66
Downloaden Sie, um offline zu lesen
1/66
Doctoral Thesis Open Defense
Development of a Lossy Online Mouse
Tracking Method for Capturing User
Interaction with Web Browser Content
Fajar Purnama D3 1
1
Human Interface Cyber Communication Laboratory (Usagawa
Lab), Computer Science and Electrical Engineering, Kumamoto
University
fajar@hicc.cs.kumamoto-u.ac.jp
July 28, 2020
2/66
Doctoral Thesis Open Defense
Chapter 1 Introduction
Chapter 1 Introduction
3/66
Doctoral Thesis Open Defense
Chapter 1 Introduction
Background
Due to Covid-19 pandemic, all activities are online.
4/66
Doctoral Thesis Open Defense
Chapter 1 Introduction
Problem
Problem 1: How to monitor/follow these activities online?
5/66
Doctoral Thesis Open Defense
Chapter 1 Introduction
Problem
Can conventional web logs / educational data only monitor
complicated online activities? Answer: Cannot
6/66
Doctoral Thesis Open Defense
Chapter 1 Introduction
Problem
Human interface that can generate deeper logs? (Solution)
conventional web logs / educational data +
7/66
Doctoral Thesis Open Defense
Chapter 1 Introduction
Problem
Why eye tracking cannot be implemented today?
Although the best monitoring quality is to use every logging and tracking there
is, eye tracking cannot be implemented today.
Eye tracking is inconvenient:
Intrusive: see previous slide where an additional goggle is necessary.
Expensive: JPY 10,000 - 1,000,000.
Mouse tracking is convenient:
No additional hardware: in default available in every computer.
Non-intrusive: users will not be aware that mouse tracking is active.
8/66
Doctoral Thesis Open Defense
Chapter 1 Introduction
Problem
Problem 2: mouse tracking applications mentioned in
academia does not suit public implementation.
OïŹ„ine Mouse Tracking Application in Laboratories:
(1) Must be installed in each client and cursor position ïŹxed to desktop.
(2) Data are stored in each client and must be retrieved individually.
Solution:
Online Mouse Tracking Client Application:
(1) Must be installed in each client but (2) data are sent to server.
Or
Online Mouse Tracking Server Application:
(1) Browser only / no additional installation and (2) data are sent to server.
9/66
Doctoral Thesis Open Defense
Chapter 1 Introduction
Problem
Problem 3: Mouse Tracking Generates Big Data Rumors
Volume: a swipe from left to right can produced 1 MB of data.
Velocity: large data generated per second (data rate).
Veracity: transmission may result in data loss due to limited connectivity.
Variety: many types of mouse and other events.
Value: the data are valuable.
Rumors are not widely investigated: if mouse tracking generates huge
data, how huge?.
10/66
Doctoral Thesis Open Defense
Chapter 1 Introduction
Objective
Objective: build application and implement.
Write an application that can capture mouse and
other browser human interfaces called mouse
tracking.
Make the mouse tracking application online that
can be implemented in public.
Investigate its data generation and resource usage.
Reduce its data generation and resource usage.
11/66
Doctoral Thesis Open Defense
Chapter 1 Introduction
Objective
Objective: dig deeper into users’ online activities.
12/66
Doctoral Thesis Open Defense
Chapter 2 Online Mouse Tracking
Chapter 2
New Online Mouse Tracking
Based on publication list, the publication that is related to this chapter is:
1* Implementation of real-time online mouse tracking on overseas quiz
session From server administrator point of view, Educational and
Information Technologies, Online First, Springer, March 2020.
13/66
Doctoral Thesis Open Defense
Chapter 2 Online Mouse Tracking
System Overview
14/66
Doctoral Thesis Open Defense
Chapter 2 Online Mouse Tracking
Mouse Tracking Desktop
Recording: clicks, movements, scrolls, zooms, keylogs, copies, and other events.
15/66
Doctoral Thesis Open Defense
Chapter 2 Online Mouse Tracking
Touch Tracking Mobile
https://hicc.cs.kumamoto-u.ac.jp/ fajar/doctoral/mouse-tracking-keyboard-
mouse-scroll-zoom-demo-and-mobile.html
16/66
Doctoral Thesis Open Defense
Chapter 2 Online Mouse Tracking
Online Mouse Tracking System
17/66
Doctoral Thesis Open Defense
Chapter 2 Online Mouse Tracking
Privacy Policy
18/66
Doctoral Thesis Open Defense
Chapter 2 Online Mouse Tracking
Implementation
Implementation
19/66
Doctoral Thesis Open Defense
Chapter 2 Online Mouse Tracking
Implementation
Client Application: Tracks All Browser’s Activities
https://github.com/0fajarpurnama0/Real-Time-Online-Mouse-Tracking-
20/66
Doctoral Thesis Open Defense
Chapter 2 Online Mouse Tracking
Implementation
Server Application: Tracks Only Website’s Activities
https://md.hicc.cs.kumamoto-u.ac.jp
21/66
Doctoral Thesis Open Defense
Chapter 2 Online Mouse Tracking
Implementation
Moodle Plugin Install
https://github.com/0fajarpurnama0/Real-Time-Online-Mouse-Tracking-
22/66
Doctoral Thesis Open Defense
Chapter 2 Online Mouse Tracking
Implementation
Mouse Tracking Quiz Session Between Mongolia and Japan
23/66
Doctoral Thesis Open Defense
Chapter 2 Online Mouse Tracking
Investigation
Investigation
24/66
Doctoral Thesis Open Defense
Chapter 2 Online Mouse Tracking
Investigation
Obtained Mouse Tracking Data
Data available: Mendeley repository titled ’Data for: Implementation of
Real-Time Online Mouse Tracking Case Study in a Small Online Quiz’.
25/66
Doctoral Thesis Open Defense
Chapter 2 Online Mouse Tracking
Investigation
Rumor and Proof of Huge Data Generation
Log File Duration Students Size
Daily Pageview City Archive 2 month - 13 kB
Moodle Log and Grades 3h 30min 41 191 kB
Moodle Log Myanmmar 6 month 41 5 MB
Mouse Tracking Mongolia 3h 30min 41 122 MB
Mouse Tracking Afganistan 3h 42 114 MB
Nasa Server Log 1995 23 days - 153 MB
Open University Learning Analytics 1 year 32593 442 MB
HarvardX Person-Course 2013 1 year 301609 33.8 MB
26/66
Doctoral Thesis Open Defense
Chapter 2 Online Mouse Tracking
Investigation
Students’ Events and Estimated Data Amount Generated
27/66
Doctoral Thesis Open Defense
Chapter 3 Resource Saving Methods
Chapter 3
Resource Saving Methods
Based on publication list, the publication that are related to this chapter are:
1* Implementation of real-time online mouse tracking on overseas quiz
session From server administrator point of view, Educational and
Information Technologies, Online First, Springer, March 2020.
2* Using Realtime Online Mouse Tracking Based on Demand
Pre-processing for Lower Storage and Transmission Cost, Journal of Big
Data, Vol.7, no.27, pp.1-22, Springer, April, 2020.
3* Demonstration on Extending The Pageview Feature to Page Section
Based: Towards Identifying Reading Patterns of Users, ICSE, November,
2016.
.
28/66
Doctoral Thesis Open Defense
Chapter 3 Resource Saving Methods
Existing Solutions
Redundant data reduction for example:
Long URL: https://www.publish0x.com/0fajarpurnama0/can-i-share-my-
work-after-copyright-transfer-xroyrle?a=4oeEw0Yb0B&tid=twitter
Into Short URL: https://t.co/Hl62T1llTV
Sampling rate (event / second) reduction for example delay capture
interval for 50 milliseconds = 20 max events per second.
Adaptive sampling: do not record during idle activities (eye tracking
record during idle activities).
Compression methods.
29/66
Doctoral Thesis Open Defense
Chapter 3 Resource Saving Methods
Not Realtime vs Realtime in Mouse Tracking
Traditionally: accumulate > wait until session ends > send all. Realtime: send
events immediately on occurrence.
30/66
Doctoral Thesis Open Defense
Chapter 3 Resource Saving Methods
Lossy Based on Demand Coordinate Compression
Default: full mouse tracking
log.
ROI: summarize coordinates
into areas.
Summarized: summarize
coordinates into amounts.
Events Summarized ROI Default
Amounts
Areas X
Coordinates X X
31/66
Doctoral Thesis Open Defense
Chapter 3 Resource Saving Methods
Default Mouse Tracking Data Transmission
Send data immediately / real time on each coordinate passed:
32/66
Doctoral Thesis Open Defense
Chapter 3 Resource Saving Methods
Region of Interest Mouse Tracking Data Transmission
Send data after leaving areas:
33/66
Doctoral Thesis Open Defense
Chapter 3 Resource Saving Methods
Summarized Event Amounts of Mouse Tracking
Send data after closing webpage:
34/66
Doctoral Thesis Open Defense
Chapter 3 Resource Saving Methods
Three Mouse Tracking Simulation
35/66
Doctoral Thesis Open Defense
Chapter 3 Resource Saving Methods
Storage Reduction
Type Queries Data Size
Default Mouse Tracking 286510 ∌100 MB
ROI Mouse Tracking 1 28048 ∌7.7 MB
ROI Mouse Tracking 2 19061 ∌5.3 MB
ROI Mouse Tracking 3 17880 ∌4.9 MB
Summarized Event Amount 23 ∌16 kB
36/66
Doctoral Thesis Open Defense
Chapter 3 Resource Saving Methods
Data Rate: Reduced > 50% storage and transmission
37/66
Doctoral Thesis Open Defense
Chapter 3 Resource Saving Methods
CPU Usage: RP3 cannot handle the default MS
38/66
Doctoral Thesis Open Defense
Chapter 3 Resource Saving Methods
RAM Usage: abnormally high for default mouse tracking
39/66
Doctoral Thesis Open Defense
Chapter 4 The depth Levels of Logs Concept
Chapter 4
The depth Levels of Logs Concept
Based on form 3, the publication that are related to this chapter are:
1* Implementation of real-time online mouse tracking on overseas quiz
session From server administrator point of view, Educational and
Information Technologies, Online First, Springer, March 2020.
2* Using Realtime Online Mouse Tracking Based on Demand
Pre-processing for Lower Storage and Transmission Cost, Journal of Big
Data, Vol.7, no.27, pp.1-22, Springer, April, 2020.
3* Demonstration on Extending The Pageview Feature to Page Section
Based: Towards Identifying Reading Patterns of Users, ICSE, November,
2016.
.
40/66
Doctoral Thesis Open Defense
Chapter 4 The depth Levels of Logs Concept
Conceptual Explaination: Log depth Level Concept
41/66
Doctoral Thesis Open Defense
Chapter 4 The depth Levels of Logs Concept
Web Page / Course Content Level Log
Web Page / Course Content Level Log
42/66
Doctoral Thesis Open Defense
Chapter 4 The depth Levels of Logs Concept
Web Page / Course Content Level Log
Mouse tracking is able to identify students leaving quiz
page (address problem 1: potential cheating).
43/66
Doctoral Thesis Open Defense
Chapter 4 The depth Levels of Logs Concept
Web Page / Course Content Level Log
Online Examination Leave Detector (Possible Software)
https://hicc.cs.kumamoto-u.ac.jp/ fajar/doctoral/mouse-tracking/exam-leave-
detect-demo.html
44/66
Doctoral Thesis Open Defense
Chapter 4 The depth Levels of Logs Concept
Area / Section Level Log
Area / Section Level Log
45/66
Doctoral Thesis Open Defense
Chapter 4 The depth Levels of Logs Concept
Area / Section Level Log
Area Heatmaps are Able to Address Problem 1: Required
Reading Passage (Proof of Read)
46/66
Doctoral Thesis Open Defense
Chapter 4 The depth Levels of Logs Concept
Area / Section Level Log
Area Heatmaps in Detail
47/66
Doctoral Thesis Open Defense
Chapter 4 The depth Levels of Logs Concept
Area / Section Level Log
Terms of Services Read Detector (Possible Software)
https://hicc.cs.kumamoto-u.ac.jp/ fajar/doctoral/mouse-tracking/eproof-of-
read.html
48/66
Doctoral Thesis Open Defense
Chapter 4 The depth Levels of Logs Concept
Coordinate / Point Level Log
Coordinate / point level log
49/66
Doctoral Thesis Open Defense
Chapter 4 The depth Levels of Logs Concept
Coordinate / Point Level Log
Clicks, Left: Triangle, Middle: Square, Right: Pentagon.
50/66
Doctoral Thesis Open Defense
conclusion and Future Work
Conclusion
Demonstrated mouse tracking as solution to problem 1 regarding to
monitoring demand for human-online interaction.
51/66
Doctoral Thesis Open Defense
conclusion and Future Work
Conclusion
Answered problem 2 regarding public mouse tracking application by
building new real-time online mouse tracking application that does not require
additional application for the client and implemented.
52/66
Doctoral Thesis Open Defense
conclusion and Future Work
Conclusion
Answered problem 3 regarding huge resource rumors by investigating
real-time online mouse tracking resource usage and reducing it over 50% by
summarizing coordinates into areas.
53/66
Doctoral Thesis Open Defense
conclusion and Future Work
Thank you
54/66
Doctoral Thesis Open Defense
Supplementary Materials
Almost All Online, Anywhere at Anytime, Face to Face
55/66
Doctoral Thesis Open Defense
Supplementary Materials
BeneïŹt and SigniïŹcance
Mouse tracking is one of the missing keys of
anything fully online.
The code is open source that can freely
implemented and developed by other parties.
The data and resource usage investigation result
can help parties to plan in implementing mouse
tracking.
Reduced data and resource usage can help those
with limited connectivity.
56/66
Doctoral Thesis Open Defense
Supplementary Materials
Obtained Moodle Log Data
57/66
Doctoral Thesis Open Defense
Supplementary Materials
Obtained Grades
58/66
Doctoral Thesis Open Defense
Supplementary Materials
Truth: Users Actually Have Full Control but Unaware
59/66
Doctoral Thesis Open Defense
Supplementary Materials
Rumor and Proof of Huge Data Generation
Moodle Log Myanmar: (CP: monmon@hicc.cs.kumamoto-u.ac.jp)
6 months of entire course with 40 students.
21098 rows of table, 5MB.
Can a swipe generate a megabyte of data?
Yes, if the width of a webpage from left to right is 1000 pixels, then a
swipe from left to right can generate at large 1000 rows of table.
Mouse Tracking Session Mongolia:
2 hours of quiz session with 22 students each hour.
286511 rows of table, 100 MB.
Mouse Tracking Session Afganistan:
2 hours of quiz session with 23 students each hour.
282601 rows of table, 92 MB.
Problem: Huge Storage and Transmission Cost
60/66
Doctoral Thesis Open Defense
Supplementary Materials
Server Storage Side: Query Amount and Data Size
Default Mouse Tracking: 100 MB
ID Name Date 2019/3/01 Duration Left Click Right Click Middle Click Mouse X Mouse Y Scroll X Scroll Y
1 Student 1 11:06:39 13.674 false false false 0 0 0 0
2 Student 1 11:06:39 0.002 false false false 1197 317 0 0
... ... ... ... ... ... ... ... ... ... ...
286511 Student 23 14:44:09 0.002 false false false 1009 179 0 0
ROI Mouse Tracking: 5.4 MB
ID Name Date 2019/3/01 Duration (second) Area (x1,x2,y1,y2) Left Clicks Right Clicks Middle Clicks Mouse Moves Scrolls
1 Student 1 11:06:39 14.148 {"header":[0,1920,0,64]} 0 0 0 1 1
2 Student 1 11:06:40 1.179 {"quiz1":[529,1900,291,570]} 0 0 0 86 0
... ... ... ... ... ... ... ... ... ...
19062 Student 23 14:44:09 0.002 {"title":[16,1904,150,270]} 0 0 0 1 0
Whole Page Summarized Mouse Tracking: 16 kB
Name Date 2019/3/01 Duration (seconds) Left Clicks Right Clicks Middle Clicks Mouse Moves Scrolls
Student 1 14:12:29 41 3 0 0 629 114
Student 2 14:44:09 90 7 0 0 1176 137
... ... ... ... ... ... ... ...
Student 22 11:55:14 2188 157 5 0 20912 6626
Student 23 11:57:37 2236 323 0 0 17982 6930
61/66
Doctoral Thesis Open Defense
Supplementary Materials
Client Side: Browser Speed
62/66
Doctoral Thesis Open Defense
Supplementary Materials
Amount of mouse event logs
Most analyzers only mined up to this level logs: page view, quiz, grades, and
other activity attempts. Even in this level, there are logs that most analyzers
are not aware of:
Amounts of events:
name duration leftclick rightclick middleclick mousemove scroll highlight
average 1368.16 175.78 0.29 7.98 11004.73 4158.12 0.61
total 56094.53 7207.00 12.00 327.00 451194.00 170483.00 25.00
minimum 41.26 3.00 0.00 0.00 633.00 116.00 0.00
maximum 2748.21 1179.00 3.00 179.00 20724.00 9521.00 4.00
stdev 728.82 295.33 0.72 39.05 4977.88 2291.82 1.16
63/66
Doctoral Thesis Open Defense
Supplementary Materials
heatmap 10 minutes interval
64/66
Doctoral Thesis Open Defense
Supplementary Materials
Grade heatmap
65/66
Doctoral Thesis Open Defense
Supplementary Materials
Why RP3? To Facilitate Those With Limited Connectivity
Hand Carry Data Collecting Through Questionnaire and Quiz Alike Using
Mini-computer Raspberry Pi, IMLF Proceeding, June, 2017.
66/66
Doctoral Thesis Open Defense
Supplementary Materials
Regular Synchronization is Necessary
From 6 Incremental Synchronization Implementation on Survey using Hand Carry Server Raspberry Pi,
IEICE Tech Report, vol.117, no.65, pp.21-24, June, 2017. and 7* Rsync and RdiïŹ€ implementation
on Moodle’s backup and restore feature for course synchronization over the network, (TENSYMP),
pp.24-29, IEEE, Mei, 2016.

Weitere Àhnliche Inhalte

Ähnlich wie Development of a lossy online mouse tracking method for capturing user interaction with web content.

Preprint-ICDMAI,Defense Institute,20-22 January 2023.pdf
Preprint-ICDMAI,Defense Institute,20-22 January 2023.pdfPreprint-ICDMAI,Defense Institute,20-22 January 2023.pdf
Preprint-ICDMAI,Defense Institute,20-22 January 2023.pdfChristo Ananth
 
Fingerprint Based Attendance System by IOT
Fingerprint Based Attendance System by IOTFingerprint Based Attendance System by IOT
Fingerprint Based Attendance System by IOTIRJET Journal
 
MedWise: Your Healthmate
MedWise: Your HealthmateMedWise: Your Healthmate
MedWise: Your HealthmateIRJET Journal
 
Application To Monitor And Manage People In Crowded Places Using Neural Networks
Application To Monitor And Manage People In Crowded Places Using Neural NetworksApplication To Monitor And Manage People In Crowded Places Using Neural Networks
Application To Monitor And Manage People In Crowded Places Using Neural NetworksIJSRED
 
Analysis of IT Monitoring Using Open Source Software Techniques: A Review
Analysis of IT Monitoring Using Open Source Software Techniques: A ReviewAnalysis of IT Monitoring Using Open Source Software Techniques: A Review
Analysis of IT Monitoring Using Open Source Software Techniques: A ReviewIJERD Editor
 
The application of process mining in a simulated smart environment to derive ...
The application of process mining in a simulated smart environment to derive ...The application of process mining in a simulated smart environment to derive ...
The application of process mining in a simulated smart environment to derive ...freedomotic
 
Real-time Classification of Malicious URLs on Twitter using Machine Activity ...
Real-time Classification of Malicious URLs on Twitter using Machine Activity ...Real-time Classification of Malicious URLs on Twitter using Machine Activity ...
Real-time Classification of Malicious URLs on Twitter using Machine Activity ...Pete Burnap
 
Template 03 - project proposal & plan
Template 03 - project proposal & planTemplate 03 - project proposal & plan
Template 03 - project proposal & planazizjaan
 
PERICLES Workflow for the automated updating of Digital Ecosystem Models with...
PERICLES Workflow for the automated updating of Digital Ecosystem Models with...PERICLES Workflow for the automated updating of Digital Ecosystem Models with...
PERICLES Workflow for the automated updating of Digital Ecosystem Models with...PERICLES_FP7
 
Social Distance Detector Using Computer Vision, OpenCV and YOLO Deep Learning...
Social Distance Detector Using Computer Vision, OpenCV and YOLO Deep Learning...Social Distance Detector Using Computer Vision, OpenCV and YOLO Deep Learning...
Social Distance Detector Using Computer Vision, OpenCV and YOLO Deep Learning...IRJET Journal
 
Major-Project2Individual (FINAL VERSION) (3)
Major-Project2Individual (FINAL VERSION) (3)Major-Project2Individual (FINAL VERSION) (3)
Major-Project2Individual (FINAL VERSION) (3)Joshua Alexander Panggabean
 
Community-based Crowdsourcing
Community-based CrowdsourcingCommunity-based Crowdsourcing
Community-based CrowdsourcingAndrea Mauri
 
OpenAIRE provide dashboard #OpenAIREweek2020
OpenAIRE provide dashboard #OpenAIREweek2020OpenAIRE provide dashboard #OpenAIREweek2020
OpenAIRE provide dashboard #OpenAIREweek2020Pedro PrĂ­ncipe
 
Trends in Information Management
Trends in Information ManagementTrends in Information Management
Trends in Information ManagementAlexander Deucalion
 
Dk tracker presentation
Dk tracker presentationDk tracker presentation
Dk tracker presentationDenaya Dennis
 
Real Time Social Distance Detector using Deep learning
Real Time Social Distance Detector using Deep learningReal Time Social Distance Detector using Deep learning
Real Time Social Distance Detector using Deep learningIRJET Journal
 
IRJET- Plagiarism Checker
IRJET- Plagiarism CheckerIRJET- Plagiarism Checker
IRJET- Plagiarism CheckerIRJET Journal
 
REMOTELY VIEW USER ACTIVITIES AND IMPOSE RULES AND PENALTIES IN A LOCAL AREA ...
REMOTELY VIEW USER ACTIVITIES AND IMPOSE RULES AND PENALTIES IN A LOCAL AREA ...REMOTELY VIEW USER ACTIVITIES AND IMPOSE RULES AND PENALTIES IN A LOCAL AREA ...
REMOTELY VIEW USER ACTIVITIES AND IMPOSE RULES AND PENALTIES IN A LOCAL AREA ...AIRCC Publishing Corporation
 

Ähnlich wie Development of a lossy online mouse tracking method for capturing user interaction with web content. (20)

Preprint-ICDMAI,Defense Institute,20-22 January 2023.pdf
Preprint-ICDMAI,Defense Institute,20-22 January 2023.pdfPreprint-ICDMAI,Defense Institute,20-22 January 2023.pdf
Preprint-ICDMAI,Defense Institute,20-22 January 2023.pdf
 
Fingerprint Based Attendance System by IOT
Fingerprint Based Attendance System by IOTFingerprint Based Attendance System by IOT
Fingerprint Based Attendance System by IOT
 
MedWise: Your Healthmate
MedWise: Your HealthmateMedWise: Your Healthmate
MedWise: Your Healthmate
 
Application To Monitor And Manage People In Crowded Places Using Neural Networks
Application To Monitor And Manage People In Crowded Places Using Neural NetworksApplication To Monitor And Manage People In Crowded Places Using Neural Networks
Application To Monitor And Manage People In Crowded Places Using Neural Networks
 
Analysis of IT Monitoring Using Open Source Software Techniques: A Review
Analysis of IT Monitoring Using Open Source Software Techniques: A ReviewAnalysis of IT Monitoring Using Open Source Software Techniques: A Review
Analysis of IT Monitoring Using Open Source Software Techniques: A Review
 
The application of process mining in a simulated smart environment to derive ...
The application of process mining in a simulated smart environment to derive ...The application of process mining in a simulated smart environment to derive ...
The application of process mining in a simulated smart environment to derive ...
 
Real-time Classification of Malicious URLs on Twitter using Machine Activity ...
Real-time Classification of Malicious URLs on Twitter using Machine Activity ...Real-time Classification of Malicious URLs on Twitter using Machine Activity ...
Real-time Classification of Malicious URLs on Twitter using Machine Activity ...
 
Template 03 - project proposal & plan
Template 03 - project proposal & planTemplate 03 - project proposal & plan
Template 03 - project proposal & plan
 
PERICLES Workflow for the automated updating of Digital Ecosystem Models with...
PERICLES Workflow for the automated updating of Digital Ecosystem Models with...PERICLES Workflow for the automated updating of Digital Ecosystem Models with...
PERICLES Workflow for the automated updating of Digital Ecosystem Models with...
 
Social Distance Detector Using Computer Vision, OpenCV and YOLO Deep Learning...
Social Distance Detector Using Computer Vision, OpenCV and YOLO Deep Learning...Social Distance Detector Using Computer Vision, OpenCV and YOLO Deep Learning...
Social Distance Detector Using Computer Vision, OpenCV and YOLO Deep Learning...
 
SyMBA: Overview
SyMBA: OverviewSyMBA: Overview
SyMBA: Overview
 
Major-Project2Individual (FINAL VERSION) (3)
Major-Project2Individual (FINAL VERSION) (3)Major-Project2Individual (FINAL VERSION) (3)
Major-Project2Individual (FINAL VERSION) (3)
 
Community-based Crowdsourcing
Community-based CrowdsourcingCommunity-based Crowdsourcing
Community-based Crowdsourcing
 
OpenAIRE provide dashboard #OpenAIREweek2020
OpenAIRE provide dashboard #OpenAIREweek2020OpenAIRE provide dashboard #OpenAIREweek2020
OpenAIRE provide dashboard #OpenAIREweek2020
 
Trends in Information Management
Trends in Information ManagementTrends in Information Management
Trends in Information Management
 
Dk tracker presentation
Dk tracker presentationDk tracker presentation
Dk tracker presentation
 
Real Time Social Distance Detector using Deep learning
Real Time Social Distance Detector using Deep learningReal Time Social Distance Detector using Deep learning
Real Time Social Distance Detector using Deep learning
 
IRJET- Plagiarism Checker
IRJET- Plagiarism CheckerIRJET- Plagiarism Checker
IRJET- Plagiarism Checker
 
Cerita
CeritaCerita
Cerita
 
REMOTELY VIEW USER ACTIVITIES AND IMPOSE RULES AND PENALTIES IN A LOCAL AREA ...
REMOTELY VIEW USER ACTIVITIES AND IMPOSE RULES AND PENALTIES IN A LOCAL AREA ...REMOTELY VIEW USER ACTIVITIES AND IMPOSE RULES AND PENALTIES IN A LOCAL AREA ...
REMOTELY VIEW USER ACTIVITIES AND IMPOSE RULES AND PENALTIES IN A LOCAL AREA ...
 

Mehr von Fajar Purnama

Nimesha Ranasinghe's Research on Digital Flavours.pptx
Nimesha Ranasinghe's Research on Digital Flavours.pptxNimesha Ranasinghe's Research on Digital Flavours.pptx
Nimesha Ranasinghe's Research on Digital Flavours.pptxFajar Purnama
 
Mata Uang Kripto 101 untuk Pengguna: BAB 1 Cerita dan Filosofi
Mata Uang Kripto 101 untuk Pengguna: BAB 1 Cerita dan FilosofiMata Uang Kripto 101 untuk Pengguna: BAB 1 Cerita dan Filosofi
Mata Uang Kripto 101 untuk Pengguna: BAB 1 Cerita dan FilosofiFajar Purnama
 
Tugas Kuliah Flow Control Error dan Selective Repeat ARQ
Tugas Kuliah Flow Control Error dan Selective Repeat ARQTugas Kuliah Flow Control Error dan Selective Repeat ARQ
Tugas Kuliah Flow Control Error dan Selective Repeat ARQFajar Purnama
 
Transmisi citra pada perangkat jaringan sensor visual nirkabel platform imote...
Transmisi citra pada perangkat jaringan sensor visual nirkabel platform imote...Transmisi citra pada perangkat jaringan sensor visual nirkabel platform imote...
Transmisi citra pada perangkat jaringan sensor visual nirkabel platform imote...Fajar Purnama
 
Recruiting future backup cryptocurrency miners with solar power bank on singl...
Recruiting future backup cryptocurrency miners with solar power bank on singl...Recruiting future backup cryptocurrency miners with solar power bank on singl...
Recruiting future backup cryptocurrency miners with solar power bank on singl...Fajar Purnama
 
fp slide on deep investigation of cross language plagiarism detection methods
fp slide on deep investigation of cross language plagiarism detection methodsfp slide on deep investigation of cross language plagiarism detection methods
fp slide on deep investigation of cross language plagiarism detection methodsFajar Purnama
 
I present someone's paper on eye tracking and web design.
I present someone's paper on eye tracking and web design.I present someone's paper on eye tracking and web design.
I present someone's paper on eye tracking and web design.Fajar Purnama
 
Portable and Synchronized Distributed Learning Management System in Severe Ne...
Portable and Synchronized Distributed Learning Management System in Severe Ne...Portable and Synchronized Distributed Learning Management System in Severe Ne...
Portable and Synchronized Distributed Learning Management System in Severe Ne...Fajar Purnama
 
Hand-Carry-Data-Collecting-Through-Questionnaire-and-Quiz-Alike-on-Crowds-usi...
Hand-Carry-Data-Collecting-Through-Questionnaire-and-Quiz-Alike-on-Crowds-usi...Hand-Carry-Data-Collecting-Through-Questionnaire-and-Quiz-Alike-on-Crowds-usi...
Hand-Carry-Data-Collecting-Through-Questionnaire-and-Quiz-Alike-on-Crowds-usi...Fajar Purnama
 
Incremental synchronization-implementation-on-survey-using-hand-carry-server-...
Incremental synchronization-implementation-on-survey-using-hand-carry-server-...Incremental synchronization-implementation-on-survey-using-hand-carry-server-...
Incremental synchronization-implementation-on-survey-using-hand-carry-server-...Fajar Purnama
 
Origin of biosignals fajar purnama 152D-8713
Origin of biosignals fajar purnama 152D-8713Origin of biosignals fajar purnama 152D-8713
Origin of biosignals fajar purnama 152D-8713Fajar Purnama
 
Demonstration on extending_the_pageview_feature_to_page_section_based_present...
Demonstration on extending_the_pageview_feature_to_page_section_based_present...Demonstration on extending_the_pageview_feature_to_page_section_based_present...
Demonstration on extending_the_pageview_feature_to_page_section_based_present...Fajar Purnama
 
Introductory Work on Section Based Page View of Web Contents Presentation
Introductory Work on Section Based Page View of Web Contents PresentationIntroductory Work on Section Based Page View of Web Contents Presentation
Introductory Work on Section Based Page View of Web Contents PresentationFajar Purnama
 
Compatible course content synchronization model for course distribution over ...
Compatible course content synchronization model for course distribution over ...Compatible course content synchronization model for course distribution over ...
Compatible course content synchronization model for course distribution over ...Fajar Purnama
 
Rdiff and Rsync Implementation on Moodle's Backup and Restore Feature of Cour...
Rdiff and Rsync Implementation on Moodle's Backup and Restore Feature of Cour...Rdiff and Rsync Implementation on Moodle's Backup and Restore Feature of Cour...
Rdiff and Rsync Implementation on Moodle's Backup and Restore Feature of Cour...Fajar Purnama
 
Fajar Purnama 152-d8713 Dynamic Content Synchronization Distributed LMS
Fajar Purnama 152-d8713 Dynamic Content Synchronization Distributed LMSFajar Purnama 152-d8713 Dynamic Content Synchronization Distributed LMS
Fajar Purnama 152-d8713 Dynamic Content Synchronization Distributed LMSFajar Purnama
 
Presentation of Understanding and Surpassing Dropbox Globecom 2015
Presentation of Understanding and Surpassing Dropbox Globecom 2015Presentation of Understanding and Surpassing Dropbox Globecom 2015
Presentation of Understanding and Surpassing Dropbox Globecom 2015Fajar Purnama
 

Mehr von Fajar Purnama (17)

Nimesha Ranasinghe's Research on Digital Flavours.pptx
Nimesha Ranasinghe's Research on Digital Flavours.pptxNimesha Ranasinghe's Research on Digital Flavours.pptx
Nimesha Ranasinghe's Research on Digital Flavours.pptx
 
Mata Uang Kripto 101 untuk Pengguna: BAB 1 Cerita dan Filosofi
Mata Uang Kripto 101 untuk Pengguna: BAB 1 Cerita dan FilosofiMata Uang Kripto 101 untuk Pengguna: BAB 1 Cerita dan Filosofi
Mata Uang Kripto 101 untuk Pengguna: BAB 1 Cerita dan Filosofi
 
Tugas Kuliah Flow Control Error dan Selective Repeat ARQ
Tugas Kuliah Flow Control Error dan Selective Repeat ARQTugas Kuliah Flow Control Error dan Selective Repeat ARQ
Tugas Kuliah Flow Control Error dan Selective Repeat ARQ
 
Transmisi citra pada perangkat jaringan sensor visual nirkabel platform imote...
Transmisi citra pada perangkat jaringan sensor visual nirkabel platform imote...Transmisi citra pada perangkat jaringan sensor visual nirkabel platform imote...
Transmisi citra pada perangkat jaringan sensor visual nirkabel platform imote...
 
Recruiting future backup cryptocurrency miners with solar power bank on singl...
Recruiting future backup cryptocurrency miners with solar power bank on singl...Recruiting future backup cryptocurrency miners with solar power bank on singl...
Recruiting future backup cryptocurrency miners with solar power bank on singl...
 
fp slide on deep investigation of cross language plagiarism detection methods
fp slide on deep investigation of cross language plagiarism detection methodsfp slide on deep investigation of cross language plagiarism detection methods
fp slide on deep investigation of cross language plagiarism detection methods
 
I present someone's paper on eye tracking and web design.
I present someone's paper on eye tracking and web design.I present someone's paper on eye tracking and web design.
I present someone's paper on eye tracking and web design.
 
Portable and Synchronized Distributed Learning Management System in Severe Ne...
Portable and Synchronized Distributed Learning Management System in Severe Ne...Portable and Synchronized Distributed Learning Management System in Severe Ne...
Portable and Synchronized Distributed Learning Management System in Severe Ne...
 
Hand-Carry-Data-Collecting-Through-Questionnaire-and-Quiz-Alike-on-Crowds-usi...
Hand-Carry-Data-Collecting-Through-Questionnaire-and-Quiz-Alike-on-Crowds-usi...Hand-Carry-Data-Collecting-Through-Questionnaire-and-Quiz-Alike-on-Crowds-usi...
Hand-Carry-Data-Collecting-Through-Questionnaire-and-Quiz-Alike-on-Crowds-usi...
 
Incremental synchronization-implementation-on-survey-using-hand-carry-server-...
Incremental synchronization-implementation-on-survey-using-hand-carry-server-...Incremental synchronization-implementation-on-survey-using-hand-carry-server-...
Incremental synchronization-implementation-on-survey-using-hand-carry-server-...
 
Origin of biosignals fajar purnama 152D-8713
Origin of biosignals fajar purnama 152D-8713Origin of biosignals fajar purnama 152D-8713
Origin of biosignals fajar purnama 152D-8713
 
Demonstration on extending_the_pageview_feature_to_page_section_based_present...
Demonstration on extending_the_pageview_feature_to_page_section_based_present...Demonstration on extending_the_pageview_feature_to_page_section_based_present...
Demonstration on extending_the_pageview_feature_to_page_section_based_present...
 
Introductory Work on Section Based Page View of Web Contents Presentation
Introductory Work on Section Based Page View of Web Contents PresentationIntroductory Work on Section Based Page View of Web Contents Presentation
Introductory Work on Section Based Page View of Web Contents Presentation
 
Compatible course content synchronization model for course distribution over ...
Compatible course content synchronization model for course distribution over ...Compatible course content synchronization model for course distribution over ...
Compatible course content synchronization model for course distribution over ...
 
Rdiff and Rsync Implementation on Moodle's Backup and Restore Feature of Cour...
Rdiff and Rsync Implementation on Moodle's Backup and Restore Feature of Cour...Rdiff and Rsync Implementation on Moodle's Backup and Restore Feature of Cour...
Rdiff and Rsync Implementation on Moodle's Backup and Restore Feature of Cour...
 
Fajar Purnama 152-d8713 Dynamic Content Synchronization Distributed LMS
Fajar Purnama 152-d8713 Dynamic Content Synchronization Distributed LMSFajar Purnama 152-d8713 Dynamic Content Synchronization Distributed LMS
Fajar Purnama 152-d8713 Dynamic Content Synchronization Distributed LMS
 
Presentation of Understanding and Surpassing Dropbox Globecom 2015
Presentation of Understanding and Surpassing Dropbox Globecom 2015Presentation of Understanding and Surpassing Dropbox Globecom 2015
Presentation of Understanding and Surpassing Dropbox Globecom 2015
 

KĂŒrzlich hochgeladen

How is AI changing journalism? (v. April 2024)
How is AI changing journalism? (v. April 2024)How is AI changing journalism? (v. April 2024)
How is AI changing journalism? (v. April 2024)Damian Radcliffe
 
CALL ON ➄8923113531 🔝Call Girls Lucknow Lucknow best sexual service Online
CALL ON ➄8923113531 🔝Call Girls Lucknow Lucknow best sexual service OnlineCALL ON ➄8923113531 🔝Call Girls Lucknow Lucknow best sexual service Online
CALL ON ➄8923113531 🔝Call Girls Lucknow Lucknow best sexual service Onlineanilsa9823
 
DDoS In Oceania and the Pacific, presented by Dave Phelan at NZNOG 2024
DDoS In Oceania and the Pacific, presented by Dave Phelan at NZNOG 2024DDoS In Oceania and the Pacific, presented by Dave Phelan at NZNOG 2024
DDoS In Oceania and the Pacific, presented by Dave Phelan at NZNOG 2024APNIC
 
Call Girls In Ashram Chowk Delhi 💯Call Us 🔝8264348440🔝
Call Girls In Ashram Chowk Delhi 💯Call Us 🔝8264348440🔝Call Girls In Ashram Chowk Delhi 💯Call Us 🔝8264348440🔝
Call Girls In Ashram Chowk Delhi 💯Call Us 🔝8264348440🔝soniya singh
 
Low Rate Young Call Girls in Sector 63 Mamura Noida ✔☆9289244007✔☆ Female E...
Low Rate Young Call Girls in Sector 63 Mamura Noida ✔☆9289244007✔☆ Female E...Low Rate Young Call Girls in Sector 63 Mamura Noida ✔☆9289244007✔☆ Female E...
Low Rate Young Call Girls in Sector 63 Mamura Noida ✔☆9289244007✔☆ Female E...SofiyaSharma5
 
đ“€€Call On 7877925207 đ“€€ Ahmedguda Call Girls Hot Model With Sexy Bhabi Ready Fo...
đ“€€Call On 7877925207 đ“€€ Ahmedguda Call Girls Hot Model With Sexy Bhabi Ready Fo...đ“€€Call On 7877925207 đ“€€ Ahmedguda Call Girls Hot Model With Sexy Bhabi Ready Fo...
đ“€€Call On 7877925207 đ“€€ Ahmedguda Call Girls Hot Model With Sexy Bhabi Ready Fo...Neha Pandey
 
(+971568250507 ))# Young Call Girls in Ajman By Pakistani Call Girls in ...
(+971568250507  ))#  Young Call Girls  in Ajman  By Pakistani Call Girls  in ...(+971568250507  ))#  Young Call Girls  in Ajman  By Pakistani Call Girls  in ...
(+971568250507 ))# Young Call Girls in Ajman By Pakistani Call Girls in ...Escorts Call Girls
 
Lucknow ❀CALL GIRL 88759*99948 ❀CALL GIRLS IN Lucknow ESCORT SERVICE❀CALL GIRL
Lucknow ❀CALL GIRL 88759*99948 ❀CALL GIRLS IN Lucknow ESCORT SERVICE❀CALL GIRLLucknow ❀CALL GIRL 88759*99948 ❀CALL GIRLS IN Lucknow ESCORT SERVICE❀CALL GIRL
Lucknow ❀CALL GIRL 88759*99948 ❀CALL GIRLS IN Lucknow ESCORT SERVICE❀CALL GIRLimonikaupta
 
WhatsApp 📞 8448380779 ✅Call Girls In Mamura Sector 66 ( Noida)
WhatsApp 📞 8448380779 ✅Call Girls In Mamura Sector 66 ( Noida)WhatsApp 📞 8448380779 ✅Call Girls In Mamura Sector 66 ( Noida)
WhatsApp 📞 8448380779 ✅Call Girls In Mamura Sector 66 ( Noida)Delhi Call girls
 
INDIVIDUAL ASSIGNMENT #3 CBG, PRESENTATION.
INDIVIDUAL ASSIGNMENT #3 CBG, PRESENTATION.INDIVIDUAL ASSIGNMENT #3 CBG, PRESENTATION.
INDIVIDUAL ASSIGNMENT #3 CBG, PRESENTATION.CarlotaBedoya1
 
GDG Cloud Southlake 32: Kyle Hettinger: Demystifying the Dark Web
GDG Cloud Southlake 32: Kyle Hettinger: Demystifying the Dark WebGDG Cloud Southlake 32: Kyle Hettinger: Demystifying the Dark Web
GDG Cloud Southlake 32: Kyle Hettinger: Demystifying the Dark WebJames Anderson
 
Delhi Call Girls Rohini 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Call
Delhi Call Girls Rohini 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip CallDelhi Call Girls Rohini 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Call
Delhi Call Girls Rohini 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Callshivangimorya083
 
Call Now ☎ 8264348440 !! Call Girls in Shahpur Jat Escort Service Delhi N.C.R.
Call Now ☎ 8264348440 !! Call Girls in Shahpur Jat Escort Service Delhi N.C.R.Call Now ☎ 8264348440 !! Call Girls in Shahpur Jat Escort Service Delhi N.C.R.
Call Now ☎ 8264348440 !! Call Girls in Shahpur Jat Escort Service Delhi N.C.R.soniya singh
 
Call Girls Dubai Prolapsed O525547819 Call Girls In Dubai Princes$
Call Girls Dubai Prolapsed O525547819 Call Girls In Dubai Princes$Call Girls Dubai Prolapsed O525547819 Call Girls In Dubai Princes$
Call Girls Dubai Prolapsed O525547819 Call Girls In Dubai Princes$kojalkojal131
 
Moving Beyond Twitter/X and Facebook - Social Media for local news providers
Moving Beyond Twitter/X and Facebook - Social Media for local news providersMoving Beyond Twitter/X and Facebook - Social Media for local news providers
Moving Beyond Twitter/X and Facebook - Social Media for local news providersDamian Radcliffe
 
On Starlink, presented by Geoff Huston at NZNOG 2024
On Starlink, presented by Geoff Huston at NZNOG 2024On Starlink, presented by Geoff Huston at NZNOG 2024
On Starlink, presented by Geoff Huston at NZNOG 2024APNIC
 

KĂŒrzlich hochgeladen (20)

How is AI changing journalism? (v. April 2024)
How is AI changing journalism? (v. April 2024)How is AI changing journalism? (v. April 2024)
How is AI changing journalism? (v. April 2024)
 
CALL ON ➄8923113531 🔝Call Girls Lucknow Lucknow best sexual service Online
CALL ON ➄8923113531 🔝Call Girls Lucknow Lucknow best sexual service OnlineCALL ON ➄8923113531 🔝Call Girls Lucknow Lucknow best sexual service Online
CALL ON ➄8923113531 🔝Call Girls Lucknow Lucknow best sexual service Online
 
DDoS In Oceania and the Pacific, presented by Dave Phelan at NZNOG 2024
DDoS In Oceania and the Pacific, presented by Dave Phelan at NZNOG 2024DDoS In Oceania and the Pacific, presented by Dave Phelan at NZNOG 2024
DDoS In Oceania and the Pacific, presented by Dave Phelan at NZNOG 2024
 
Call Girls In Ashram Chowk Delhi 💯Call Us 🔝8264348440🔝
Call Girls In Ashram Chowk Delhi 💯Call Us 🔝8264348440🔝Call Girls In Ashram Chowk Delhi 💯Call Us 🔝8264348440🔝
Call Girls In Ashram Chowk Delhi 💯Call Us 🔝8264348440🔝
 
Low Rate Young Call Girls in Sector 63 Mamura Noida ✔☆9289244007✔☆ Female E...
Low Rate Young Call Girls in Sector 63 Mamura Noida ✔☆9289244007✔☆ Female E...Low Rate Young Call Girls in Sector 63 Mamura Noida ✔☆9289244007✔☆ Female E...
Low Rate Young Call Girls in Sector 63 Mamura Noida ✔☆9289244007✔☆ Female E...
 
Rohini Sector 6 Call Girls Delhi 9999965857 @Sabina Saikh No Advance
Rohini Sector 6 Call Girls Delhi 9999965857 @Sabina Saikh No AdvanceRohini Sector 6 Call Girls Delhi 9999965857 @Sabina Saikh No Advance
Rohini Sector 6 Call Girls Delhi 9999965857 @Sabina Saikh No Advance
 
đ“€€Call On 7877925207 đ“€€ Ahmedguda Call Girls Hot Model With Sexy Bhabi Ready Fo...
đ“€€Call On 7877925207 đ“€€ Ahmedguda Call Girls Hot Model With Sexy Bhabi Ready Fo...đ“€€Call On 7877925207 đ“€€ Ahmedguda Call Girls Hot Model With Sexy Bhabi Ready Fo...
đ“€€Call On 7877925207 đ“€€ Ahmedguda Call Girls Hot Model With Sexy Bhabi Ready Fo...
 
(+971568250507 ))# Young Call Girls in Ajman By Pakistani Call Girls in ...
(+971568250507  ))#  Young Call Girls  in Ajman  By Pakistani Call Girls  in ...(+971568250507  ))#  Young Call Girls  in Ajman  By Pakistani Call Girls  in ...
(+971568250507 ))# Young Call Girls in Ajman By Pakistani Call Girls in ...
 
Lucknow ❀CALL GIRL 88759*99948 ❀CALL GIRLS IN Lucknow ESCORT SERVICE❀CALL GIRL
Lucknow ❀CALL GIRL 88759*99948 ❀CALL GIRLS IN Lucknow ESCORT SERVICE❀CALL GIRLLucknow ❀CALL GIRL 88759*99948 ❀CALL GIRLS IN Lucknow ESCORT SERVICE❀CALL GIRL
Lucknow ❀CALL GIRL 88759*99948 ❀CALL GIRLS IN Lucknow ESCORT SERVICE❀CALL GIRL
 
WhatsApp 📞 8448380779 ✅Call Girls In Mamura Sector 66 ( Noida)
WhatsApp 📞 8448380779 ✅Call Girls In Mamura Sector 66 ( Noida)WhatsApp 📞 8448380779 ✅Call Girls In Mamura Sector 66 ( Noida)
WhatsApp 📞 8448380779 ✅Call Girls In Mamura Sector 66 ( Noida)
 
INDIVIDUAL ASSIGNMENT #3 CBG, PRESENTATION.
INDIVIDUAL ASSIGNMENT #3 CBG, PRESENTATION.INDIVIDUAL ASSIGNMENT #3 CBG, PRESENTATION.
INDIVIDUAL ASSIGNMENT #3 CBG, PRESENTATION.
 
Rohini Sector 26 Call Girls Delhi 9999965857 @Sabina Saikh No Advance
Rohini Sector 26 Call Girls Delhi 9999965857 @Sabina Saikh No AdvanceRohini Sector 26 Call Girls Delhi 9999965857 @Sabina Saikh No Advance
Rohini Sector 26 Call Girls Delhi 9999965857 @Sabina Saikh No Advance
 
GDG Cloud Southlake 32: Kyle Hettinger: Demystifying the Dark Web
GDG Cloud Southlake 32: Kyle Hettinger: Demystifying the Dark WebGDG Cloud Southlake 32: Kyle Hettinger: Demystifying the Dark Web
GDG Cloud Southlake 32: Kyle Hettinger: Demystifying the Dark Web
 
Delhi Call Girls Rohini 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Call
Delhi Call Girls Rohini 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip CallDelhi Call Girls Rohini 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Call
Delhi Call Girls Rohini 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Call
 
Call Now ☎ 8264348440 !! Call Girls in Shahpur Jat Escort Service Delhi N.C.R.
Call Now ☎ 8264348440 !! Call Girls in Shahpur Jat Escort Service Delhi N.C.R.Call Now ☎ 8264348440 !! Call Girls in Shahpur Jat Escort Service Delhi N.C.R.
Call Now ☎ 8264348440 !! Call Girls in Shahpur Jat Escort Service Delhi N.C.R.
 
Call Girls Dubai Prolapsed O525547819 Call Girls In Dubai Princes$
Call Girls Dubai Prolapsed O525547819 Call Girls In Dubai Princes$Call Girls Dubai Prolapsed O525547819 Call Girls In Dubai Princes$
Call Girls Dubai Prolapsed O525547819 Call Girls In Dubai Princes$
 
(INDIRA) Call Girl Pune Call Now 8250077686 Pune Escorts 24x7
(INDIRA) Call Girl Pune Call Now 8250077686 Pune Escorts 24x7(INDIRA) Call Girl Pune Call Now 8250077686 Pune Escorts 24x7
(INDIRA) Call Girl Pune Call Now 8250077686 Pune Escorts 24x7
 
Moving Beyond Twitter/X and Facebook - Social Media for local news providers
Moving Beyond Twitter/X and Facebook - Social Media for local news providersMoving Beyond Twitter/X and Facebook - Social Media for local news providers
Moving Beyond Twitter/X and Facebook - Social Media for local news providers
 
@9999965857 đŸ«Š Sexy Desi Call Girls Laxmi Nagar 💓 High Profile Escorts Delhi đŸ«¶
@9999965857 đŸ«Š Sexy Desi Call Girls Laxmi Nagar 💓 High Profile Escorts Delhi đŸ«¶@9999965857 đŸ«Š Sexy Desi Call Girls Laxmi Nagar 💓 High Profile Escorts Delhi đŸ«¶
@9999965857 đŸ«Š Sexy Desi Call Girls Laxmi Nagar 💓 High Profile Escorts Delhi đŸ«¶
 
On Starlink, presented by Geoff Huston at NZNOG 2024
On Starlink, presented by Geoff Huston at NZNOG 2024On Starlink, presented by Geoff Huston at NZNOG 2024
On Starlink, presented by Geoff Huston at NZNOG 2024
 

Development of a lossy online mouse tracking method for capturing user interaction with web content.

  • 1. 1/66 Doctoral Thesis Open Defense Development of a Lossy Online Mouse Tracking Method for Capturing User Interaction with Web Browser Content Fajar Purnama D3 1 1 Human Interface Cyber Communication Laboratory (Usagawa Lab), Computer Science and Electrical Engineering, Kumamoto University fajar@hicc.cs.kumamoto-u.ac.jp July 28, 2020
  • 2. 2/66 Doctoral Thesis Open Defense Chapter 1 Introduction Chapter 1 Introduction
  • 3. 3/66 Doctoral Thesis Open Defense Chapter 1 Introduction Background Due to Covid-19 pandemic, all activities are online.
  • 4. 4/66 Doctoral Thesis Open Defense Chapter 1 Introduction Problem Problem 1: How to monitor/follow these activities online?
  • 5. 5/66 Doctoral Thesis Open Defense Chapter 1 Introduction Problem Can conventional web logs / educational data only monitor complicated online activities? Answer: Cannot
  • 6. 6/66 Doctoral Thesis Open Defense Chapter 1 Introduction Problem Human interface that can generate deeper logs? (Solution) conventional web logs / educational data +
  • 7. 7/66 Doctoral Thesis Open Defense Chapter 1 Introduction Problem Why eye tracking cannot be implemented today? Although the best monitoring quality is to use every logging and tracking there is, eye tracking cannot be implemented today. Eye tracking is inconvenient: Intrusive: see previous slide where an additional goggle is necessary. Expensive: JPY 10,000 - 1,000,000. Mouse tracking is convenient: No additional hardware: in default available in every computer. Non-intrusive: users will not be aware that mouse tracking is active.
  • 8. 8/66 Doctoral Thesis Open Defense Chapter 1 Introduction Problem Problem 2: mouse tracking applications mentioned in academia does not suit public implementation. OïŹ„ine Mouse Tracking Application in Laboratories: (1) Must be installed in each client and cursor position ïŹxed to desktop. (2) Data are stored in each client and must be retrieved individually. Solution: Online Mouse Tracking Client Application: (1) Must be installed in each client but (2) data are sent to server. Or Online Mouse Tracking Server Application: (1) Browser only / no additional installation and (2) data are sent to server.
  • 9. 9/66 Doctoral Thesis Open Defense Chapter 1 Introduction Problem Problem 3: Mouse Tracking Generates Big Data Rumors Volume: a swipe from left to right can produced 1 MB of data. Velocity: large data generated per second (data rate). Veracity: transmission may result in data loss due to limited connectivity. Variety: many types of mouse and other events. Value: the data are valuable. Rumors are not widely investigated: if mouse tracking generates huge data, how huge?.
  • 10. 10/66 Doctoral Thesis Open Defense Chapter 1 Introduction Objective Objective: build application and implement. Write an application that can capture mouse and other browser human interfaces called mouse tracking. Make the mouse tracking application online that can be implemented in public. Investigate its data generation and resource usage. Reduce its data generation and resource usage.
  • 11. 11/66 Doctoral Thesis Open Defense Chapter 1 Introduction Objective Objective: dig deeper into users’ online activities.
  • 12. 12/66 Doctoral Thesis Open Defense Chapter 2 Online Mouse Tracking Chapter 2 New Online Mouse Tracking Based on publication list, the publication that is related to this chapter is: 1* Implementation of real-time online mouse tracking on overseas quiz session From server administrator point of view, Educational and Information Technologies, Online First, Springer, March 2020.
  • 13. 13/66 Doctoral Thesis Open Defense Chapter 2 Online Mouse Tracking System Overview
  • 14. 14/66 Doctoral Thesis Open Defense Chapter 2 Online Mouse Tracking Mouse Tracking Desktop Recording: clicks, movements, scrolls, zooms, keylogs, copies, and other events.
  • 15. 15/66 Doctoral Thesis Open Defense Chapter 2 Online Mouse Tracking Touch Tracking Mobile https://hicc.cs.kumamoto-u.ac.jp/ fajar/doctoral/mouse-tracking-keyboard- mouse-scroll-zoom-demo-and-mobile.html
  • 16. 16/66 Doctoral Thesis Open Defense Chapter 2 Online Mouse Tracking Online Mouse Tracking System
  • 17. 17/66 Doctoral Thesis Open Defense Chapter 2 Online Mouse Tracking Privacy Policy
  • 18. 18/66 Doctoral Thesis Open Defense Chapter 2 Online Mouse Tracking Implementation Implementation
  • 19. 19/66 Doctoral Thesis Open Defense Chapter 2 Online Mouse Tracking Implementation Client Application: Tracks All Browser’s Activities https://github.com/0fajarpurnama0/Real-Time-Online-Mouse-Tracking-
  • 20. 20/66 Doctoral Thesis Open Defense Chapter 2 Online Mouse Tracking Implementation Server Application: Tracks Only Website’s Activities https://md.hicc.cs.kumamoto-u.ac.jp
  • 21. 21/66 Doctoral Thesis Open Defense Chapter 2 Online Mouse Tracking Implementation Moodle Plugin Install https://github.com/0fajarpurnama0/Real-Time-Online-Mouse-Tracking-
  • 22. 22/66 Doctoral Thesis Open Defense Chapter 2 Online Mouse Tracking Implementation Mouse Tracking Quiz Session Between Mongolia and Japan
  • 23. 23/66 Doctoral Thesis Open Defense Chapter 2 Online Mouse Tracking Investigation Investigation
  • 24. 24/66 Doctoral Thesis Open Defense Chapter 2 Online Mouse Tracking Investigation Obtained Mouse Tracking Data Data available: Mendeley repository titled ’Data for: Implementation of Real-Time Online Mouse Tracking Case Study in a Small Online Quiz’.
  • 25. 25/66 Doctoral Thesis Open Defense Chapter 2 Online Mouse Tracking Investigation Rumor and Proof of Huge Data Generation Log File Duration Students Size Daily Pageview City Archive 2 month - 13 kB Moodle Log and Grades 3h 30min 41 191 kB Moodle Log Myanmmar 6 month 41 5 MB Mouse Tracking Mongolia 3h 30min 41 122 MB Mouse Tracking Afganistan 3h 42 114 MB Nasa Server Log 1995 23 days - 153 MB Open University Learning Analytics 1 year 32593 442 MB HarvardX Person-Course 2013 1 year 301609 33.8 MB
  • 26. 26/66 Doctoral Thesis Open Defense Chapter 2 Online Mouse Tracking Investigation Students’ Events and Estimated Data Amount Generated
  • 27. 27/66 Doctoral Thesis Open Defense Chapter 3 Resource Saving Methods Chapter 3 Resource Saving Methods Based on publication list, the publication that are related to this chapter are: 1* Implementation of real-time online mouse tracking on overseas quiz session From server administrator point of view, Educational and Information Technologies, Online First, Springer, March 2020. 2* Using Realtime Online Mouse Tracking Based on Demand Pre-processing for Lower Storage and Transmission Cost, Journal of Big Data, Vol.7, no.27, pp.1-22, Springer, April, 2020. 3* Demonstration on Extending The Pageview Feature to Page Section Based: Towards Identifying Reading Patterns of Users, ICSE, November, 2016. .
  • 28. 28/66 Doctoral Thesis Open Defense Chapter 3 Resource Saving Methods Existing Solutions Redundant data reduction for example: Long URL: https://www.publish0x.com/0fajarpurnama0/can-i-share-my- work-after-copyright-transfer-xroyrle?a=4oeEw0Yb0B&tid=twitter Into Short URL: https://t.co/Hl62T1llTV Sampling rate (event / second) reduction for example delay capture interval for 50 milliseconds = 20 max events per second. Adaptive sampling: do not record during idle activities (eye tracking record during idle activities). Compression methods.
  • 29. 29/66 Doctoral Thesis Open Defense Chapter 3 Resource Saving Methods Not Realtime vs Realtime in Mouse Tracking Traditionally: accumulate > wait until session ends > send all. Realtime: send events immediately on occurrence.
  • 30. 30/66 Doctoral Thesis Open Defense Chapter 3 Resource Saving Methods Lossy Based on Demand Coordinate Compression Default: full mouse tracking log. ROI: summarize coordinates into areas. Summarized: summarize coordinates into amounts. Events Summarized ROI Default Amounts Areas X Coordinates X X
  • 31. 31/66 Doctoral Thesis Open Defense Chapter 3 Resource Saving Methods Default Mouse Tracking Data Transmission Send data immediately / real time on each coordinate passed:
  • 32. 32/66 Doctoral Thesis Open Defense Chapter 3 Resource Saving Methods Region of Interest Mouse Tracking Data Transmission Send data after leaving areas:
  • 33. 33/66 Doctoral Thesis Open Defense Chapter 3 Resource Saving Methods Summarized Event Amounts of Mouse Tracking Send data after closing webpage:
  • 34. 34/66 Doctoral Thesis Open Defense Chapter 3 Resource Saving Methods Three Mouse Tracking Simulation
  • 35. 35/66 Doctoral Thesis Open Defense Chapter 3 Resource Saving Methods Storage Reduction Type Queries Data Size Default Mouse Tracking 286510 ∌100 MB ROI Mouse Tracking 1 28048 ∌7.7 MB ROI Mouse Tracking 2 19061 ∌5.3 MB ROI Mouse Tracking 3 17880 ∌4.9 MB Summarized Event Amount 23 ∌16 kB
  • 36. 36/66 Doctoral Thesis Open Defense Chapter 3 Resource Saving Methods Data Rate: Reduced > 50% storage and transmission
  • 37. 37/66 Doctoral Thesis Open Defense Chapter 3 Resource Saving Methods CPU Usage: RP3 cannot handle the default MS
  • 38. 38/66 Doctoral Thesis Open Defense Chapter 3 Resource Saving Methods RAM Usage: abnormally high for default mouse tracking
  • 39. 39/66 Doctoral Thesis Open Defense Chapter 4 The depth Levels of Logs Concept Chapter 4 The depth Levels of Logs Concept Based on form 3, the publication that are related to this chapter are: 1* Implementation of real-time online mouse tracking on overseas quiz session From server administrator point of view, Educational and Information Technologies, Online First, Springer, March 2020. 2* Using Realtime Online Mouse Tracking Based on Demand Pre-processing for Lower Storage and Transmission Cost, Journal of Big Data, Vol.7, no.27, pp.1-22, Springer, April, 2020. 3* Demonstration on Extending The Pageview Feature to Page Section Based: Towards Identifying Reading Patterns of Users, ICSE, November, 2016. .
  • 40. 40/66 Doctoral Thesis Open Defense Chapter 4 The depth Levels of Logs Concept Conceptual Explaination: Log depth Level Concept
  • 41. 41/66 Doctoral Thesis Open Defense Chapter 4 The depth Levels of Logs Concept Web Page / Course Content Level Log Web Page / Course Content Level Log
  • 42. 42/66 Doctoral Thesis Open Defense Chapter 4 The depth Levels of Logs Concept Web Page / Course Content Level Log Mouse tracking is able to identify students leaving quiz page (address problem 1: potential cheating).
  • 43. 43/66 Doctoral Thesis Open Defense Chapter 4 The depth Levels of Logs Concept Web Page / Course Content Level Log Online Examination Leave Detector (Possible Software) https://hicc.cs.kumamoto-u.ac.jp/ fajar/doctoral/mouse-tracking/exam-leave- detect-demo.html
  • 44. 44/66 Doctoral Thesis Open Defense Chapter 4 The depth Levels of Logs Concept Area / Section Level Log Area / Section Level Log
  • 45. 45/66 Doctoral Thesis Open Defense Chapter 4 The depth Levels of Logs Concept Area / Section Level Log Area Heatmaps are Able to Address Problem 1: Required Reading Passage (Proof of Read)
  • 46. 46/66 Doctoral Thesis Open Defense Chapter 4 The depth Levels of Logs Concept Area / Section Level Log Area Heatmaps in Detail
  • 47. 47/66 Doctoral Thesis Open Defense Chapter 4 The depth Levels of Logs Concept Area / Section Level Log Terms of Services Read Detector (Possible Software) https://hicc.cs.kumamoto-u.ac.jp/ fajar/doctoral/mouse-tracking/eproof-of- read.html
  • 48. 48/66 Doctoral Thesis Open Defense Chapter 4 The depth Levels of Logs Concept Coordinate / Point Level Log Coordinate / point level log
  • 49. 49/66 Doctoral Thesis Open Defense Chapter 4 The depth Levels of Logs Concept Coordinate / Point Level Log Clicks, Left: Triangle, Middle: Square, Right: Pentagon.
  • 50. 50/66 Doctoral Thesis Open Defense conclusion and Future Work Conclusion Demonstrated mouse tracking as solution to problem 1 regarding to monitoring demand for human-online interaction.
  • 51. 51/66 Doctoral Thesis Open Defense conclusion and Future Work Conclusion Answered problem 2 regarding public mouse tracking application by building new real-time online mouse tracking application that does not require additional application for the client and implemented.
  • 52. 52/66 Doctoral Thesis Open Defense conclusion and Future Work Conclusion Answered problem 3 regarding huge resource rumors by investigating real-time online mouse tracking resource usage and reducing it over 50% by summarizing coordinates into areas.
  • 53. 53/66 Doctoral Thesis Open Defense conclusion and Future Work Thank you
  • 54. 54/66 Doctoral Thesis Open Defense Supplementary Materials Almost All Online, Anywhere at Anytime, Face to Face
  • 55. 55/66 Doctoral Thesis Open Defense Supplementary Materials BeneïŹt and SigniïŹcance Mouse tracking is one of the missing keys of anything fully online. The code is open source that can freely implemented and developed by other parties. The data and resource usage investigation result can help parties to plan in implementing mouse tracking. Reduced data and resource usage can help those with limited connectivity.
  • 56. 56/66 Doctoral Thesis Open Defense Supplementary Materials Obtained Moodle Log Data
  • 57. 57/66 Doctoral Thesis Open Defense Supplementary Materials Obtained Grades
  • 58. 58/66 Doctoral Thesis Open Defense Supplementary Materials Truth: Users Actually Have Full Control but Unaware
  • 59. 59/66 Doctoral Thesis Open Defense Supplementary Materials Rumor and Proof of Huge Data Generation Moodle Log Myanmar: (CP: monmon@hicc.cs.kumamoto-u.ac.jp) 6 months of entire course with 40 students. 21098 rows of table, 5MB. Can a swipe generate a megabyte of data? Yes, if the width of a webpage from left to right is 1000 pixels, then a swipe from left to right can generate at large 1000 rows of table. Mouse Tracking Session Mongolia: 2 hours of quiz session with 22 students each hour. 286511 rows of table, 100 MB. Mouse Tracking Session Afganistan: 2 hours of quiz session with 23 students each hour. 282601 rows of table, 92 MB. Problem: Huge Storage and Transmission Cost
  • 60. 60/66 Doctoral Thesis Open Defense Supplementary Materials Server Storage Side: Query Amount and Data Size Default Mouse Tracking: 100 MB ID Name Date 2019/3/01 Duration Left Click Right Click Middle Click Mouse X Mouse Y Scroll X Scroll Y 1 Student 1 11:06:39 13.674 false false false 0 0 0 0 2 Student 1 11:06:39 0.002 false false false 1197 317 0 0 ... ... ... ... ... ... ... ... ... ... ... 286511 Student 23 14:44:09 0.002 false false false 1009 179 0 0 ROI Mouse Tracking: 5.4 MB ID Name Date 2019/3/01 Duration (second) Area (x1,x2,y1,y2) Left Clicks Right Clicks Middle Clicks Mouse Moves Scrolls 1 Student 1 11:06:39 14.148 {"header":[0,1920,0,64]} 0 0 0 1 1 2 Student 1 11:06:40 1.179 {"quiz1":[529,1900,291,570]} 0 0 0 86 0 ... ... ... ... ... ... ... ... ... ... 19062 Student 23 14:44:09 0.002 {"title":[16,1904,150,270]} 0 0 0 1 0 Whole Page Summarized Mouse Tracking: 16 kB Name Date 2019/3/01 Duration (seconds) Left Clicks Right Clicks Middle Clicks Mouse Moves Scrolls Student 1 14:12:29 41 3 0 0 629 114 Student 2 14:44:09 90 7 0 0 1176 137 ... ... ... ... ... ... ... ... Student 22 11:55:14 2188 157 5 0 20912 6626 Student 23 11:57:37 2236 323 0 0 17982 6930
  • 61. 61/66 Doctoral Thesis Open Defense Supplementary Materials Client Side: Browser Speed
  • 62. 62/66 Doctoral Thesis Open Defense Supplementary Materials Amount of mouse event logs Most analyzers only mined up to this level logs: page view, quiz, grades, and other activity attempts. Even in this level, there are logs that most analyzers are not aware of: Amounts of events: name duration leftclick rightclick middleclick mousemove scroll highlight average 1368.16 175.78 0.29 7.98 11004.73 4158.12 0.61 total 56094.53 7207.00 12.00 327.00 451194.00 170483.00 25.00 minimum 41.26 3.00 0.00 0.00 633.00 116.00 0.00 maximum 2748.21 1179.00 3.00 179.00 20724.00 9521.00 4.00 stdev 728.82 295.33 0.72 39.05 4977.88 2291.82 1.16
  • 63. 63/66 Doctoral Thesis Open Defense Supplementary Materials heatmap 10 minutes interval
  • 64. 64/66 Doctoral Thesis Open Defense Supplementary Materials Grade heatmap
  • 65. 65/66 Doctoral Thesis Open Defense Supplementary Materials Why RP3? To Facilitate Those With Limited Connectivity Hand Carry Data Collecting Through Questionnaire and Quiz Alike Using Mini-computer Raspberry Pi, IMLF Proceeding, June, 2017.
  • 66. 66/66 Doctoral Thesis Open Defense Supplementary Materials Regular Synchronization is Necessary From 6 Incremental Synchronization Implementation on Survey using Hand Carry Server Raspberry Pi, IEICE Tech Report, vol.117, no.65, pp.21-24, June, 2017. and 7* Rsync and RdiïŹ€ implementation on Moodle’s backup and restore feature for course synchronization over the network, (TENSYMP), pp.24-29, IEEE, Mei, 2016.