SlideShare a Scribd company logo
1 of 24
Download to read offline
1
CLASSROOM OCCUPANCY PROJECT
2
PROJECT OBJECTIVES MEET OUR TEAM
N I KO L AY B A N D U R A
K R I S T E N M C I N T Y R E
A B R A H A M M O N T I L L A
M E N G D I Y U E
S V E T L A N A Z O LOTA R E VA
Use Raspberry Pi 3 and other
devices to capture sensor data from
classroom
CA PTURE SEN SO R DATA
Build and tune Supervised
Classification and Regression
Models
M ACHIN E L EA RN IN G M O DEL S
Create a web application that
incorporates supervised machine
learning models to predict real-time
occupancy levels
B UILD W EB A PPLICATIO N
3
INITIAL SETUP
Original Raspberry Pi 3 setup:
 Bluetooth Devices
 Door Sensor
 Motion Sensor
 Dual Temperature & Humidity Sensor
Ultimately, the Motion Sensor was removed
and the following sensors were added:
 Camera
 CO2
 Light
 Noise
D A T A I N G E S T I O N
4
FINAL SETUP
Raspberry Pi 3 sensors are
connected to a breadboard.
SENSORS
BLUETOOTH DEVICES
CO2, ppm
DOOR SENSOR
HUMIDITY, %
IMAGE, 8-MP
LIGHT, Lux
NOISE, Hz
TEMPERATURE, °Celsius
D A T A I N G E S T I O N
5
L E A R N M O R E
6
DATA
STORAGE
D A T A I N G E S T I O N
7
FLASK APP
OCCUPANCY LOGGING
APPLICATION
DATA STORAGE
D A T A I N G E S T I O N
8
CAMERA SETUP
Saturday, June 10, 2017
D A T A I N G E S T I O N
9
IMAGE DATA
Histogram of a Picture (vector)
D A T A I N G E S T I O N
10
IMAGE DATA
Difference of Adjacent Picture Histograms
D A T A I N G E S T I O N
11
EXPLORATORY
DATA ANALYSIS
STEP 1
OUTLIER
DETECTION
STEP 2 RESAMPLE
DATASET
STEP 3
FEATURE
SELECTION
STEP 4
DATA WRANGLING PROCESS
12
DAILY CO2 DATA
D A T A W R A N G L I N E
CO2 SPIKE
D A T A W R A N G L I N G
Saturday, April 1, 2017
13
DAILY LIGHT DATA
D A T A W R A N G L I N G
Initial Histogram of Light Data
14
BLUETOOTH DEVICES DATA
Pearson score
Missing Data
W R I T E H E R E
B l u e t o o t h D e v i c e s : F r i d a y , M a y 5 , 2 0 1 7
D A T A W R A N G L I N G
15
FEATURE CORRELATION
A company is an association or
collection of individuals, whether
natural persons, legal
W R I T E H E R E
B l u e t o o t h D e v i c e s & N o N - P e r s o n a l B l u e t o o t h D e v i c e s
D A T A W R A N G L I N G
16
MACHINE LEARNING
M A I N C H A L L E N G E S
Mi s s i n g Val u e s
 S e n so r E rro rs
 N ew Fe at u re s
T i m e -Se ri e s D ata
 Tim e S e rie sS p lit
 C A RT M o d e ls
Cl as s I m b al an ce
 8 9 % Oc c u p ie d
 Oc c u p a n c y C ate go r y
 0 : E mpt y
 1 -1 6 : L ow
 1 7 - 2 7 : M id - L eve l
 > 2 7 : Hig h
17
MODEL
CLASSIFICATION
REPORT
CROSS-
VALIDATION
ACCURACY
SCORES
GaussianNB f1 score: 0.98
precision: 0.98
recall: 0.98
0.8056 Training set: 0.962
Test set: 0.983
kNN f1 score: 0.50
precision: 0.49
recall: 0.58
0.5755 Training set: 0.922
Test set: 0.583
LDA f1 score: 0.96
precision: 0.96
recall: 0.96
0.8699 Training set: 0.896
Test set: 0.960
Logistic
Regression
f1 score: 0.94
precision: 0.95
recall: 0.94
0.8067 Training set: 0.913
Test set: 0.942
SGD f1 score: 0.42
precision: 0.33
recall: 0.55
0.6223 Training set: 0.661
Test set: 0.574
SVC f1 score: 0.64
precision: 0.81
recall: 0.70
0.6402 Training set: 0.991
Test set: 0.703
CLASSIFICATION
MODELS
M A C H I N E L E A R N I N G
C L A S S I F I C AT I O N
R E P O R T
C R O S S - VA L I D AT I O N
S C O R E ( 1 2 - F o l d )
A C C U R A C Y S C O R E S
INITIAL RESULTS
18
NAÏVE BAYES MODEL
M A C H I N E L E A R N I N G
19
kNN Class Balance
M A C H I N E L E A R N I N G
kNN Elbow Plot
M A C H I N E L E A R N I N G
20
MODEL
CLASSIFICATION
REPORT
BEST SCORE
MULTI-CLASS F1-
SCORE
SCALER TUNED PARAMETERS
kNN f1 score: 0.81
precision: 0.84
recall: 0.83
0.7300 micro: 0.8255
weighted: 0.8149
macro: 0.7635
RobustScaler n_neighbors: 4
Logistic
Regression
f1 score: 0.95
precision: 0.97
recall: 0.96
0.8858 micro: 0.9565
weighted: 0.9548
macro: 0.9145
N/A C: 110
class_weight: balanced
SGD f1 score: 0.91
precision: 0.94
recall: 0.92
0.8568 micro: 0.9188
weighted: 0.9109
macro: 0.8280
StandardScaler alpha: 0.001
n_iter: 5
penalty: l1
SVC f1 score: 1.00
precision: 1.00
recall: 1.00
0.9320 micro: 0.9971
weighted: 0.9971
macro: 0.9942
RobustScaler kernel: linear
C: 10
class_weight: balanced
CLASSIFICATION MODELS
TUNED RESULTS
M A C H I N E L E A R N I N G
21
CLASS BALANCE PLOTCLASSIFICATION REPORT
TUNED SVC MODEL
S V C M o d e l S V C M o d e l
M A C H I N E L E A R N I N G
22
PREDICTION ERRORRESIDUALS PLOT
REGRESSION MODELS
L I N E A R R E G R E S S I O N L I N E A R R E G R E S S I O N
M A C H I N E L E A R N I N G
23
INITIAL STATE
DASHBOARD
W E B A P P L I C A T I O N
An interactive viewer
that allows users to
mine data and collect
targeted statistics
WAV E S
Set an action to occur
for a specified condition
in the data stream
T R I G G E R S
Each tile is customized
to display a unique
event stream
T I L E S
24
MODEL OPTIMATIZATION
C L A S S R O O M O C C U P A N C Y C A P S T O N E
 Specific to location
 User privacy
L I M I TAT I O N S
 Gather sensor data from multiple rooms at
the same location
 Take into account the building’s HVAC
system
 Gather data from rooms impacted by
outdoor weather conditions
I M P R OV E M E N T S

More Related Content

Similar to Classroom Occupancy Machine Learning Project

AUTO AI 2021 talk Real world data augmentations for autonomous driving : B Ra...
AUTO AI 2021 talk Real world data augmentations for autonomous driving : B Ra...AUTO AI 2021 talk Real world data augmentations for autonomous driving : B Ra...
AUTO AI 2021 talk Real world data augmentations for autonomous driving : B Ra...Ravi Kiran B.
 
Medical Image Segmentation Using Hidden Markov Random Field A Distributed Ap...
Medical Image Segmentation Using Hidden Markov Random Field  A Distributed Ap...Medical Image Segmentation Using Hidden Markov Random Field  A Distributed Ap...
Medical Image Segmentation Using Hidden Markov Random Field A Distributed Ap...EL-Hachemi Guerrout
 
Optical fiber communication prof harsha sanap (1)
Optical fiber communication  prof harsha sanap (1)Optical fiber communication  prof harsha sanap (1)
Optical fiber communication prof harsha sanap (1)xavier engg college mahim
 
Model Serving via Pulsar Functions
Model Serving via Pulsar FunctionsModel Serving via Pulsar Functions
Model Serving via Pulsar FunctionsArun Kejariwal
 
Comaprision of s7 hp and tmw2100 tx
Comaprision of s7 hp and tmw2100 txComaprision of s7 hp and tmw2100 tx
Comaprision of s7 hp and tmw2100 txR.Narasimha Swamy
 
What every C++ programmer should know about modern compilers (w/o comments, A...
What every C++ programmer should know about modern compilers (w/o comments, A...What every C++ programmer should know about modern compilers (w/o comments, A...
What every C++ programmer should know about modern compilers (w/o comments, A...Sławomir Zborowski
 
Vu_HPSC2012_02.pptx
Vu_HPSC2012_02.pptxVu_HPSC2012_02.pptx
Vu_HPSC2012_02.pptxQucngV
 
8 inch TFT-LCD Datesheet, AUO, 800*1280, MIPI Interface
8 inch TFT-LCD Datesheet, AUO, 800*1280, MIPI Interface8 inch TFT-LCD Datesheet, AUO, 800*1280, MIPI Interface
8 inch TFT-LCD Datesheet, AUO, 800*1280, MIPI InterfacePanox Display
 
Compact Street Lights - 25W LED STELLAR STREET LIGHT Specifications
Compact Street Lights - 25W LED STELLAR STREET LIGHT SpecificationsCompact Street Lights - 25W LED STELLAR STREET LIGHT Specifications
Compact Street Lights - 25W LED STELLAR STREET LIGHT SpecificationsCompact Lighting
 
Problem 7PurposeBreak apart a complicated system.ConstantsC7C13.docx
Problem 7PurposeBreak apart a complicated system.ConstantsC7C13.docxProblem 7PurposeBreak apart a complicated system.ConstantsC7C13.docx
Problem 7PurposeBreak apart a complicated system.ConstantsC7C13.docxLacieKlineeb
 
Thomas Roth-Berghofer (University of West London) – Artificial Intelligence -...
Thomas Roth-Berghofer (University of West London) – Artificial Intelligence -...Thomas Roth-Berghofer (University of West London) – Artificial Intelligence -...
Thomas Roth-Berghofer (University of West London) – Artificial Intelligence -...Techsylvania
 
Artificial Intelligence – Case-based reasoning for recommender systems – Invi...
Artificial Intelligence – Case-based reasoning for recommender systems – Invi...Artificial Intelligence – Case-based reasoning for recommender systems – Invi...
Artificial Intelligence – Case-based reasoning for recommender systems – Invi...Thomas Roth-Berghofer
 
Design and Development of Automatic Water Dispenser
Design and Development of Automatic Water DispenserDesign and Development of Automatic Water Dispenser
Design and Development of Automatic Water DispenserMuhammad Saif Ul Islam
 
Experiences from Designing and Validating a Software Modernization Transforma...
Experiences from Designing and Validating a Software Modernization Transforma...Experiences from Designing and Validating a Software Modernization Transforma...
Experiences from Designing and Validating a Software Modernization Transforma...Alexandru-Florin Iosif-Lazăr
 

Similar to Classroom Occupancy Machine Learning Project (20)

AUTO AI 2021 talk Real world data augmentations for autonomous driving : B Ra...
AUTO AI 2021 talk Real world data augmentations for autonomous driving : B Ra...AUTO AI 2021 talk Real world data augmentations for autonomous driving : B Ra...
AUTO AI 2021 talk Real world data augmentations for autonomous driving : B Ra...
 
Medical Image Segmentation Using Hidden Markov Random Field A Distributed Ap...
Medical Image Segmentation Using Hidden Markov Random Field  A Distributed Ap...Medical Image Segmentation Using Hidden Markov Random Field  A Distributed Ap...
Medical Image Segmentation Using Hidden Markov Random Field A Distributed Ap...
 
Optical fiber communication prof harsha sanap (1)
Optical fiber communication  prof harsha sanap (1)Optical fiber communication  prof harsha sanap (1)
Optical fiber communication prof harsha sanap (1)
 
Practica1 digi2
Practica1 digi2Practica1 digi2
Practica1 digi2
 
Lecture 3 data-driven models
Lecture 3 data-driven modelsLecture 3 data-driven models
Lecture 3 data-driven models
 
2014 PV Performance Modeling Workshop: Outdoor Module Characterization Method...
2014 PV Performance Modeling Workshop: Outdoor Module Characterization Method...2014 PV Performance Modeling Workshop: Outdoor Module Characterization Method...
2014 PV Performance Modeling Workshop: Outdoor Module Characterization Method...
 
Model Serving via Pulsar Functions
Model Serving via Pulsar FunctionsModel Serving via Pulsar Functions
Model Serving via Pulsar Functions
 
Machine learning
Machine learningMachine learning
Machine learning
 
Comaprision of s7 hp and tmw2100 tx
Comaprision of s7 hp and tmw2100 txComaprision of s7 hp and tmw2100 tx
Comaprision of s7 hp and tmw2100 tx
 
What every C++ programmer should know about modern compilers (w/o comments, A...
What every C++ programmer should know about modern compilers (w/o comments, A...What every C++ programmer should know about modern compilers (w/o comments, A...
What every C++ programmer should know about modern compilers (w/o comments, A...
 
Vu_HPSC2012_02.pptx
Vu_HPSC2012_02.pptxVu_HPSC2012_02.pptx
Vu_HPSC2012_02.pptx
 
8 inch TFT-LCD Datesheet, AUO, 800*1280, MIPI Interface
8 inch TFT-LCD Datesheet, AUO, 800*1280, MIPI Interface8 inch TFT-LCD Datesheet, AUO, 800*1280, MIPI Interface
8 inch TFT-LCD Datesheet, AUO, 800*1280, MIPI Interface
 
Compact Street Lights - 25W LED STELLAR STREET LIGHT Specifications
Compact Street Lights - 25W LED STELLAR STREET LIGHT SpecificationsCompact Street Lights - 25W LED STELLAR STREET LIGHT Specifications
Compact Street Lights - 25W LED STELLAR STREET LIGHT Specifications
 
Wcre12c.ppt
Wcre12c.pptWcre12c.ppt
Wcre12c.ppt
 
Problem 7PurposeBreak apart a complicated system.ConstantsC7C13.docx
Problem 7PurposeBreak apart a complicated system.ConstantsC7C13.docxProblem 7PurposeBreak apart a complicated system.ConstantsC7C13.docx
Problem 7PurposeBreak apart a complicated system.ConstantsC7C13.docx
 
Thomas Roth-Berghofer (University of West London) – Artificial Intelligence -...
Thomas Roth-Berghofer (University of West London) – Artificial Intelligence -...Thomas Roth-Berghofer (University of West London) – Artificial Intelligence -...
Thomas Roth-Berghofer (University of West London) – Artificial Intelligence -...
 
Artificial Intelligence – Case-based reasoning for recommender systems – Invi...
Artificial Intelligence – Case-based reasoning for recommender systems – Invi...Artificial Intelligence – Case-based reasoning for recommender systems – Invi...
Artificial Intelligence – Case-based reasoning for recommender systems – Invi...
 
Design and Development of Automatic Water Dispenser
Design and Development of Automatic Water DispenserDesign and Development of Automatic Water Dispenser
Design and Development of Automatic Water Dispenser
 
Experiences from Designing and Validating a Software Modernization Transforma...
Experiences from Designing and Validating a Software Modernization Transforma...Experiences from Designing and Validating a Software Modernization Transforma...
Experiences from Designing and Validating a Software Modernization Transforma...
 
Ica bada
Ica badaIca bada
Ica bada
 

Recently uploaded

Predictive Precipitation: Advanced Rain Forecasting Techniques
Predictive Precipitation: Advanced Rain Forecasting TechniquesPredictive Precipitation: Advanced Rain Forecasting Techniques
Predictive Precipitation: Advanced Rain Forecasting TechniquesBoston Institute of Analytics
 
Genuine love spell caster )! ,+27834335081) Ex lover back permanently in At...
Genuine love spell caster )! ,+27834335081)   Ex lover back permanently in At...Genuine love spell caster )! ,+27834335081)   Ex lover back permanently in At...
Genuine love spell caster )! ,+27834335081) Ex lover back permanently in At...BabaJohn3
 
Data Visualization Exploring and Explaining with Data 1st Edition by Camm sol...
Data Visualization Exploring and Explaining with Data 1st Edition by Camm sol...Data Visualization Exploring and Explaining with Data 1st Edition by Camm sol...
Data Visualization Exploring and Explaining with Data 1st Edition by Camm sol...ssuserf63bd7
 
Digital Marketing Demystified: Expert Tips from Samantha Rae Coolbeth
Digital Marketing Demystified: Expert Tips from Samantha Rae CoolbethDigital Marketing Demystified: Expert Tips from Samantha Rae Coolbeth
Digital Marketing Demystified: Expert Tips from Samantha Rae CoolbethSamantha Rae Coolbeth
 
NO1 Best Kala Jadu Expert Specialist In Germany Kala Jadu Expert Specialist I...
NO1 Best Kala Jadu Expert Specialist In Germany Kala Jadu Expert Specialist I...NO1 Best Kala Jadu Expert Specialist In Germany Kala Jadu Expert Specialist I...
NO1 Best Kala Jadu Expert Specialist In Germany Kala Jadu Expert Specialist I...Amil baba
 
如何办理加州大学伯克利分校毕业证(UCB毕业证)成绩单留信学历认证
如何办理加州大学伯克利分校毕业证(UCB毕业证)成绩单留信学历认证如何办理加州大学伯克利分校毕业证(UCB毕业证)成绩单留信学历认证
如何办理加州大学伯克利分校毕业证(UCB毕业证)成绩单留信学历认证a8om7o51
 
原件一样伦敦国王学院毕业证成绩单留信学历认证
原件一样伦敦国王学院毕业证成绩单留信学历认证原件一样伦敦国王学院毕业证成绩单留信学历认证
原件一样伦敦国王学院毕业证成绩单留信学历认证pwgnohujw
 
1:1原版定制利物浦大学毕业证(Liverpool毕业证)成绩单学位证书留信学历认证
1:1原版定制利物浦大学毕业证(Liverpool毕业证)成绩单学位证书留信学历认证1:1原版定制利物浦大学毕业证(Liverpool毕业证)成绩单学位证书留信学历认证
1:1原版定制利物浦大学毕业证(Liverpool毕业证)成绩单学位证书留信学历认证ppy8zfkfm
 
如何办理(Dalhousie毕业证书)达尔豪斯大学毕业证成绩单留信学历认证
如何办理(Dalhousie毕业证书)达尔豪斯大学毕业证成绩单留信学历认证如何办理(Dalhousie毕业证书)达尔豪斯大学毕业证成绩单留信学历认证
如何办理(Dalhousie毕业证书)达尔豪斯大学毕业证成绩单留信学历认证zifhagzkk
 
Northern New England Tableau User Group (TUG) May 2024
Northern New England Tableau User Group (TUG) May 2024Northern New England Tableau User Group (TUG) May 2024
Northern New England Tableau User Group (TUG) May 2024patrickdtherriault
 
如何办理英国卡迪夫大学毕业证(Cardiff毕业证书)成绩单留信学历认证
如何办理英国卡迪夫大学毕业证(Cardiff毕业证书)成绩单留信学历认证如何办理英国卡迪夫大学毕业证(Cardiff毕业证书)成绩单留信学历认证
如何办理英国卡迪夫大学毕业证(Cardiff毕业证书)成绩单留信学历认证ju0dztxtn
 
SCI8-Q4-MOD11.pdfwrwujrrjfaajerjrajrrarj
SCI8-Q4-MOD11.pdfwrwujrrjfaajerjrajrrarjSCI8-Q4-MOD11.pdfwrwujrrjfaajerjrajrrarj
SCI8-Q4-MOD11.pdfwrwujrrjfaajerjrajrrarjadimosmejiaslendon
 
Audience Researchndfhcvnfgvgbhujhgfv.pptx
Audience Researchndfhcvnfgvgbhujhgfv.pptxAudience Researchndfhcvnfgvgbhujhgfv.pptx
Audience Researchndfhcvnfgvgbhujhgfv.pptxStephen266013
 
如何办理(WashU毕业证书)圣路易斯华盛顿大学毕业证成绩单本科硕士学位证留信学历认证
如何办理(WashU毕业证书)圣路易斯华盛顿大学毕业证成绩单本科硕士学位证留信学历认证如何办理(WashU毕业证书)圣路易斯华盛顿大学毕业证成绩单本科硕士学位证留信学历认证
如何办理(WashU毕业证书)圣路易斯华盛顿大学毕业证成绩单本科硕士学位证留信学历认证acoha1
 
原件一样(UWO毕业证书)西安大略大学毕业证成绩单留信学历认证
原件一样(UWO毕业证书)西安大略大学毕业证成绩单留信学历认证原件一样(UWO毕业证书)西安大略大学毕业证成绩单留信学历认证
原件一样(UWO毕业证书)西安大略大学毕业证成绩单留信学历认证pwgnohujw
 
What is Insertion Sort. Its basic information
What is Insertion Sort. Its basic informationWhat is Insertion Sort. Its basic information
What is Insertion Sort. Its basic informationmuqadasqasim10
 
Seven tools of quality control.slideshare
Seven tools of quality control.slideshareSeven tools of quality control.slideshare
Seven tools of quality control.slideshareraiaryan448
 
obat aborsi Bontang wa 081336238223 jual obat aborsi cytotec asli di Bontang6...
obat aborsi Bontang wa 081336238223 jual obat aborsi cytotec asli di Bontang6...obat aborsi Bontang wa 081336238223 jual obat aborsi cytotec asli di Bontang6...
obat aborsi Bontang wa 081336238223 jual obat aborsi cytotec asli di Bontang6...yulianti213969
 
Data Analytics for Digital Marketing Lecture for Advanced Digital & Social Me...
Data Analytics for Digital Marketing Lecture for Advanced Digital & Social Me...Data Analytics for Digital Marketing Lecture for Advanced Digital & Social Me...
Data Analytics for Digital Marketing Lecture for Advanced Digital & Social Me...Valters Lauzums
 
Aggregations - The Elasticsearch "GROUP BY"
Aggregations - The Elasticsearch "GROUP BY"Aggregations - The Elasticsearch "GROUP BY"
Aggregations - The Elasticsearch "GROUP BY"John Sobanski
 

Recently uploaded (20)

Predictive Precipitation: Advanced Rain Forecasting Techniques
Predictive Precipitation: Advanced Rain Forecasting TechniquesPredictive Precipitation: Advanced Rain Forecasting Techniques
Predictive Precipitation: Advanced Rain Forecasting Techniques
 
Genuine love spell caster )! ,+27834335081) Ex lover back permanently in At...
Genuine love spell caster )! ,+27834335081)   Ex lover back permanently in At...Genuine love spell caster )! ,+27834335081)   Ex lover back permanently in At...
Genuine love spell caster )! ,+27834335081) Ex lover back permanently in At...
 
Data Visualization Exploring and Explaining with Data 1st Edition by Camm sol...
Data Visualization Exploring and Explaining with Data 1st Edition by Camm sol...Data Visualization Exploring and Explaining with Data 1st Edition by Camm sol...
Data Visualization Exploring and Explaining with Data 1st Edition by Camm sol...
 
Digital Marketing Demystified: Expert Tips from Samantha Rae Coolbeth
Digital Marketing Demystified: Expert Tips from Samantha Rae CoolbethDigital Marketing Demystified: Expert Tips from Samantha Rae Coolbeth
Digital Marketing Demystified: Expert Tips from Samantha Rae Coolbeth
 
NO1 Best Kala Jadu Expert Specialist In Germany Kala Jadu Expert Specialist I...
NO1 Best Kala Jadu Expert Specialist In Germany Kala Jadu Expert Specialist I...NO1 Best Kala Jadu Expert Specialist In Germany Kala Jadu Expert Specialist I...
NO1 Best Kala Jadu Expert Specialist In Germany Kala Jadu Expert Specialist I...
 
如何办理加州大学伯克利分校毕业证(UCB毕业证)成绩单留信学历认证
如何办理加州大学伯克利分校毕业证(UCB毕业证)成绩单留信学历认证如何办理加州大学伯克利分校毕业证(UCB毕业证)成绩单留信学历认证
如何办理加州大学伯克利分校毕业证(UCB毕业证)成绩单留信学历认证
 
原件一样伦敦国王学院毕业证成绩单留信学历认证
原件一样伦敦国王学院毕业证成绩单留信学历认证原件一样伦敦国王学院毕业证成绩单留信学历认证
原件一样伦敦国王学院毕业证成绩单留信学历认证
 
1:1原版定制利物浦大学毕业证(Liverpool毕业证)成绩单学位证书留信学历认证
1:1原版定制利物浦大学毕业证(Liverpool毕业证)成绩单学位证书留信学历认证1:1原版定制利物浦大学毕业证(Liverpool毕业证)成绩单学位证书留信学历认证
1:1原版定制利物浦大学毕业证(Liverpool毕业证)成绩单学位证书留信学历认证
 
如何办理(Dalhousie毕业证书)达尔豪斯大学毕业证成绩单留信学历认证
如何办理(Dalhousie毕业证书)达尔豪斯大学毕业证成绩单留信学历认证如何办理(Dalhousie毕业证书)达尔豪斯大学毕业证成绩单留信学历认证
如何办理(Dalhousie毕业证书)达尔豪斯大学毕业证成绩单留信学历认证
 
Northern New England Tableau User Group (TUG) May 2024
Northern New England Tableau User Group (TUG) May 2024Northern New England Tableau User Group (TUG) May 2024
Northern New England Tableau User Group (TUG) May 2024
 
如何办理英国卡迪夫大学毕业证(Cardiff毕业证书)成绩单留信学历认证
如何办理英国卡迪夫大学毕业证(Cardiff毕业证书)成绩单留信学历认证如何办理英国卡迪夫大学毕业证(Cardiff毕业证书)成绩单留信学历认证
如何办理英国卡迪夫大学毕业证(Cardiff毕业证书)成绩单留信学历认证
 
SCI8-Q4-MOD11.pdfwrwujrrjfaajerjrajrrarj
SCI8-Q4-MOD11.pdfwrwujrrjfaajerjrajrrarjSCI8-Q4-MOD11.pdfwrwujrrjfaajerjrajrrarj
SCI8-Q4-MOD11.pdfwrwujrrjfaajerjrajrrarj
 
Audience Researchndfhcvnfgvgbhujhgfv.pptx
Audience Researchndfhcvnfgvgbhujhgfv.pptxAudience Researchndfhcvnfgvgbhujhgfv.pptx
Audience Researchndfhcvnfgvgbhujhgfv.pptx
 
如何办理(WashU毕业证书)圣路易斯华盛顿大学毕业证成绩单本科硕士学位证留信学历认证
如何办理(WashU毕业证书)圣路易斯华盛顿大学毕业证成绩单本科硕士学位证留信学历认证如何办理(WashU毕业证书)圣路易斯华盛顿大学毕业证成绩单本科硕士学位证留信学历认证
如何办理(WashU毕业证书)圣路易斯华盛顿大学毕业证成绩单本科硕士学位证留信学历认证
 
原件一样(UWO毕业证书)西安大略大学毕业证成绩单留信学历认证
原件一样(UWO毕业证书)西安大略大学毕业证成绩单留信学历认证原件一样(UWO毕业证书)西安大略大学毕业证成绩单留信学历认证
原件一样(UWO毕业证书)西安大略大学毕业证成绩单留信学历认证
 
What is Insertion Sort. Its basic information
What is Insertion Sort. Its basic informationWhat is Insertion Sort. Its basic information
What is Insertion Sort. Its basic information
 
Seven tools of quality control.slideshare
Seven tools of quality control.slideshareSeven tools of quality control.slideshare
Seven tools of quality control.slideshare
 
obat aborsi Bontang wa 081336238223 jual obat aborsi cytotec asli di Bontang6...
obat aborsi Bontang wa 081336238223 jual obat aborsi cytotec asli di Bontang6...obat aborsi Bontang wa 081336238223 jual obat aborsi cytotec asli di Bontang6...
obat aborsi Bontang wa 081336238223 jual obat aborsi cytotec asli di Bontang6...
 
Data Analytics for Digital Marketing Lecture for Advanced Digital & Social Me...
Data Analytics for Digital Marketing Lecture for Advanced Digital & Social Me...Data Analytics for Digital Marketing Lecture for Advanced Digital & Social Me...
Data Analytics for Digital Marketing Lecture for Advanced Digital & Social Me...
 
Aggregations - The Elasticsearch "GROUP BY"
Aggregations - The Elasticsearch "GROUP BY"Aggregations - The Elasticsearch "GROUP BY"
Aggregations - The Elasticsearch "GROUP BY"
 

Classroom Occupancy Machine Learning Project

  • 2. 2 PROJECT OBJECTIVES MEET OUR TEAM N I KO L AY B A N D U R A K R I S T E N M C I N T Y R E A B R A H A M M O N T I L L A M E N G D I Y U E S V E T L A N A Z O LOTA R E VA Use Raspberry Pi 3 and other devices to capture sensor data from classroom CA PTURE SEN SO R DATA Build and tune Supervised Classification and Regression Models M ACHIN E L EA RN IN G M O DEL S Create a web application that incorporates supervised machine learning models to predict real-time occupancy levels B UILD W EB A PPLICATIO N
  • 3. 3 INITIAL SETUP Original Raspberry Pi 3 setup:  Bluetooth Devices  Door Sensor  Motion Sensor  Dual Temperature & Humidity Sensor Ultimately, the Motion Sensor was removed and the following sensors were added:  Camera  CO2  Light  Noise D A T A I N G E S T I O N
  • 4. 4 FINAL SETUP Raspberry Pi 3 sensors are connected to a breadboard. SENSORS BLUETOOTH DEVICES CO2, ppm DOOR SENSOR HUMIDITY, % IMAGE, 8-MP LIGHT, Lux NOISE, Hz TEMPERATURE, °Celsius D A T A I N G E S T I O N
  • 5. 5 L E A R N M O R E
  • 6. 6 DATA STORAGE D A T A I N G E S T I O N
  • 7. 7 FLASK APP OCCUPANCY LOGGING APPLICATION DATA STORAGE D A T A I N G E S T I O N
  • 8. 8 CAMERA SETUP Saturday, June 10, 2017 D A T A I N G E S T I O N
  • 9. 9 IMAGE DATA Histogram of a Picture (vector) D A T A I N G E S T I O N
  • 10. 10 IMAGE DATA Difference of Adjacent Picture Histograms D A T A I N G E S T I O N
  • 11. 11 EXPLORATORY DATA ANALYSIS STEP 1 OUTLIER DETECTION STEP 2 RESAMPLE DATASET STEP 3 FEATURE SELECTION STEP 4 DATA WRANGLING PROCESS
  • 12. 12 DAILY CO2 DATA D A T A W R A N G L I N E CO2 SPIKE D A T A W R A N G L I N G Saturday, April 1, 2017
  • 13. 13 DAILY LIGHT DATA D A T A W R A N G L I N G Initial Histogram of Light Data
  • 14. 14 BLUETOOTH DEVICES DATA Pearson score Missing Data W R I T E H E R E B l u e t o o t h D e v i c e s : F r i d a y , M a y 5 , 2 0 1 7 D A T A W R A N G L I N G
  • 15. 15 FEATURE CORRELATION A company is an association or collection of individuals, whether natural persons, legal W R I T E H E R E B l u e t o o t h D e v i c e s & N o N - P e r s o n a l B l u e t o o t h D e v i c e s D A T A W R A N G L I N G
  • 16. 16 MACHINE LEARNING M A I N C H A L L E N G E S Mi s s i n g Val u e s  S e n so r E rro rs  N ew Fe at u re s T i m e -Se ri e s D ata  Tim e S e rie sS p lit  C A RT M o d e ls Cl as s I m b al an ce  8 9 % Oc c u p ie d  Oc c u p a n c y C ate go r y  0 : E mpt y  1 -1 6 : L ow  1 7 - 2 7 : M id - L eve l  > 2 7 : Hig h
  • 17. 17 MODEL CLASSIFICATION REPORT CROSS- VALIDATION ACCURACY SCORES GaussianNB f1 score: 0.98 precision: 0.98 recall: 0.98 0.8056 Training set: 0.962 Test set: 0.983 kNN f1 score: 0.50 precision: 0.49 recall: 0.58 0.5755 Training set: 0.922 Test set: 0.583 LDA f1 score: 0.96 precision: 0.96 recall: 0.96 0.8699 Training set: 0.896 Test set: 0.960 Logistic Regression f1 score: 0.94 precision: 0.95 recall: 0.94 0.8067 Training set: 0.913 Test set: 0.942 SGD f1 score: 0.42 precision: 0.33 recall: 0.55 0.6223 Training set: 0.661 Test set: 0.574 SVC f1 score: 0.64 precision: 0.81 recall: 0.70 0.6402 Training set: 0.991 Test set: 0.703 CLASSIFICATION MODELS M A C H I N E L E A R N I N G C L A S S I F I C AT I O N R E P O R T C R O S S - VA L I D AT I O N S C O R E ( 1 2 - F o l d ) A C C U R A C Y S C O R E S INITIAL RESULTS
  • 18. 18 NAÏVE BAYES MODEL M A C H I N E L E A R N I N G
  • 19. 19 kNN Class Balance M A C H I N E L E A R N I N G kNN Elbow Plot M A C H I N E L E A R N I N G
  • 20. 20 MODEL CLASSIFICATION REPORT BEST SCORE MULTI-CLASS F1- SCORE SCALER TUNED PARAMETERS kNN f1 score: 0.81 precision: 0.84 recall: 0.83 0.7300 micro: 0.8255 weighted: 0.8149 macro: 0.7635 RobustScaler n_neighbors: 4 Logistic Regression f1 score: 0.95 precision: 0.97 recall: 0.96 0.8858 micro: 0.9565 weighted: 0.9548 macro: 0.9145 N/A C: 110 class_weight: balanced SGD f1 score: 0.91 precision: 0.94 recall: 0.92 0.8568 micro: 0.9188 weighted: 0.9109 macro: 0.8280 StandardScaler alpha: 0.001 n_iter: 5 penalty: l1 SVC f1 score: 1.00 precision: 1.00 recall: 1.00 0.9320 micro: 0.9971 weighted: 0.9971 macro: 0.9942 RobustScaler kernel: linear C: 10 class_weight: balanced CLASSIFICATION MODELS TUNED RESULTS M A C H I N E L E A R N I N G
  • 21. 21 CLASS BALANCE PLOTCLASSIFICATION REPORT TUNED SVC MODEL S V C M o d e l S V C M o d e l M A C H I N E L E A R N I N G
  • 22. 22 PREDICTION ERRORRESIDUALS PLOT REGRESSION MODELS L I N E A R R E G R E S S I O N L I N E A R R E G R E S S I O N M A C H I N E L E A R N I N G
  • 23. 23 INITIAL STATE DASHBOARD W E B A P P L I C A T I O N An interactive viewer that allows users to mine data and collect targeted statistics WAV E S Set an action to occur for a specified condition in the data stream T R I G G E R S Each tile is customized to display a unique event stream T I L E S
  • 24. 24 MODEL OPTIMATIZATION C L A S S R O O M O C C U P A N C Y C A P S T O N E  Specific to location  User privacy L I M I TAT I O N S  Gather sensor data from multiple rooms at the same location  Take into account the building’s HVAC system  Gather data from rooms impacted by outdoor weather conditions I M P R OV E M E N T S