SlideShare a Scribd company logo
1 of 38
Download to read offline
1
2018 2018/10/7
1.
2.
3.IT
4.
5.
6.
7.
(@TeitoNakagawa)
/
GIS
2018
IT
IT
Dev
OpsData
• Smartphone App
• Agile Development
• Device
• Cloud
• Help Desk
• Data Science
• Analytics
*http://www.itmedia.co.jp/enterprise/articles/
1802/26/news007.html
*http://diamond.jp/articles/-/150122
20172015 20162014
Legacy
SendMail
H/W
Analog
Feature Phone
W/F
In-House
DevOps
Digital
Phase 1
Cloud
Phase 0
On-Premises
(1/2)
20182017
CRM/SFA
MA
D-Marketing
IoT
Collab.
Phase 3
Insighti
Real Estate SPA
DD-Biz.
ML/DL
Big Data
PoC
Phase 2
Agility
(2/2)
• 2015
Web
• ! " #
→
IT
Dev
OpsData
•
•
•
•
•
1.
2.
3.
Data
BIGQuery
DB GIS
Web
SFA embulk
Web
SFA
RDB
BQ
BQGIS
GIS
.shp
( F LVZ
• KPSJQ C8B G
• 1 G 1 G 1
• O N , C 4
• C8B G )
• aW I
• JTI CG 7
BigQuery
–
–
–
–
•
•
•
–
–
–
–
•
–
–
–2m
2m
3m
• or
• or or Web
…
Q. A.
–
–
Q. or A.
–
–
A
B
C
D
3:7 2:2
→
Stacking (Acc:88%)
10
(Acc:86%)
(Acc:82%)
!
"
"
!
!
•
: .. /// . /
) ( 0
Comparison of urban land
price prediction approaches
Data : introduction
● Public Land Price data released by the Ministry of Land, Infrastructure, Transport and
Tourism as of 2018/01/01
● Area Tokyo, Nagoya, Kanagawa, Chiba and Saitama (without Tokyo Islands)
● Type of lands Residential
Model 1 : Geographically Weighted Regression (GWR)
“Everything is related to everything else. But near things are more related than distant things”
Tobler’s first law of Geography
Normal Regression
Every point is treated the same for prediction
GWR
Closer points are treated as more important :
the closer the bigger the weight
10
30
30
10
20
10
30
30
10
20
Prediction:
20
Prediction:
12
● Model definition:
with yi the land price, xik the value of the variable k, (ui,vi) the coordinates, βk the regression
parameter for the variable k and εi the error at location i
● Parameter estimation (regression) in matrix notation:
with W(ui,vi) the diagonal matrix denoting the geographical weighting of each observed data
for regression point i
Model 1 : Geographically Weighted Regression (GWR)
● Weight at regression point i of datapoint j :
● Bandwidth (scale) selection with a Golden-Section algorithm
● Bandwidth is adaptive rather than fixed :
Adaptive bandwidth that includes the k-nearest neighbors at each regression point rather
than a fixed value
This allows to incorporate data points with few close neighbors better in the regression
Model 1 : Geographically Weighted Regression (GWR)
where dij is the distance between i and
j, and b is the bandwidth or scale
Model 2 : Hedonic
● Model definition :
● Regression method: OLS
Yi Land Price in /m2
(log)
Variables (xk) Units Explanations
Distance to closest station m (log)
Distance to big railway hubs m (log) ( : : ...)
Floor-area ratio %
Road width m
Gas Flag : connected to gas infrastructure or not
Land area m² (log)
Building material Categorical concrete, wood,...
Land usage Categorical low-rise residential type-1, semi-residential, ...
Neighborhood weighted prices distance-weighted average of the 9 nearest neighbors
Model 3 : Boosted Trees (XGBoost)
● A set of weak learners (decision trees) are combined to get strong learners
● Trees are grown sequentially : each tree is grown using information from
previously grown trees
● Boosted trees are implemented using XGBoost library
● Same variables as Model 2 (Hedonic) Distance to closest station, Distance
to big railway hubs, Floor-area ratio, Road width, Gas, Land area, Building
material, Land usage, Neighborhood weighted prices
ERRORS
TRAIN TEST TRAIN TEST
MODEL MODEL
DATASET
Results : Ratios of low-error predictions
Error Global Tokyo Saitama Kanagawa Chiba Nagoya
GWR
< 5 % 32 % 39 % 27 % 34 % 22 % 34 %
< 10% 57 % 63 % 57 % 62 % 37 % 59 %
< 20% 83 % 86 % 85 % 89 % 67 % 87 %
Hedonic
< 5 % 34 % 38 % 32 % 33 % 21 % 34%
< 10% 56 % 63 % 64 % 57 % 41 % 59 %
< 20% 83 % 85 % 89 % 85 % 64 % 87 %
XGBoost
< 5 % 35 % 41 % 34 % 35 % 21 % 35 %
< 10% 61 % 66 % 66 % 63 % 42 % 61 %
< 20% 85 % 92 % 88 % 88 % 69 % 88 %
Results : prediction error distribution
GWR 11.9 % 8.4 %
Hedonic 11.9 % 8.7 %
XGBoost 11.3 % 8.3 %
● Error estimation via a 100-fold Cross Validation
(train / eval to data ratio : 75% / 25%)
Mean prediction error by area Error distribution for each model
Results : maps
Predicted prices map around Tokyo Prediction error around Tokyo
Results : discussion
XGBoost was the best performing model
● Model Limitations and possible ameliorations
○ GWR
The current version is single scale, assuming that all the variables experience local effects on the
same scale.
Multiscale GWR would drop that assumption and potentially improve the accuracy
○ Hedonic/XGBoost :
Spatial correlation effects are treated empirically
● Data :
○ Data points : some areas have little data point, dragging the overall accuracy down (especially Chiba)
○ Euclidean distance is used for in each case
Other distance metrics (Manhattan distance, commute time,..) might be more suited
• IBIS 2018 2018 11
• /
オープンハウスにおける機械学習・データサイエンスの取り組みについて

More Related Content

What's hot

3D Analyst - Watershed, Padang
3D Analyst - Watershed, Padang3D Analyst - Watershed, Padang
3D Analyst - Watershed, PadangHartanto Sanjaya
 
DSD-INT 2018 iMOD version 4.3 double precision big coordinates - Vermeulen
DSD-INT 2018 iMOD version 4.3 double precision big coordinates - VermeulenDSD-INT 2018 iMOD version 4.3 double precision big coordinates - Vermeulen
DSD-INT 2018 iMOD version 4.3 double precision big coordinates - VermeulenDeltares
 
3D Analyst - Watershed, Lombok
3D Analyst - Watershed, Lombok3D Analyst - Watershed, Lombok
3D Analyst - Watershed, LombokHartanto Sanjaya
 
A Visualization Application for SIAT
A Visualization Application for SIATA Visualization Application for SIAT
A Visualization Application for SIATBernhard Snizek
 
Preparing LiDAR for Use in ArcGIS 10.1 with the Data Interoperability Extension
Preparing LiDAR for Use in ArcGIS 10.1 with the Data Interoperability ExtensionPreparing LiDAR for Use in ArcGIS 10.1 with the Data Interoperability Extension
Preparing LiDAR for Use in ArcGIS 10.1 with the Data Interoperability ExtensionSafe Software
 
2017 Vendor Showcase Track: Tracking Z: Limitations of the World We Live In
2017 Vendor Showcase Track:  Tracking Z: Limitations of the World We Live In2017 Vendor Showcase Track:  Tracking Z: Limitations of the World We Live In
2017 Vendor Showcase Track: Tracking Z: Limitations of the World We Live InGIS in the Rockies
 
3d hydrogeological conceptual model building in denmark
3d hydrogeological conceptual model building in denmark3d hydrogeological conceptual model building in denmark
3d hydrogeological conceptual model building in denmarkTorben Bach
 
E Cognition User Summit2009 C Storch Gaf Emlc
E Cognition User Summit2009 C Storch Gaf EmlcE Cognition User Summit2009 C Storch Gaf Emlc
E Cognition User Summit2009 C Storch Gaf EmlcTrimble Geospatial Munich
 
Smart Interpretation - Fast AEM Modelling - SAGEEP 2017
Smart Interpretation - Fast AEM Modelling - SAGEEP 2017Smart Interpretation - Fast AEM Modelling - SAGEEP 2017
Smart Interpretation - Fast AEM Modelling - SAGEEP 2017Torben Bach
 
How Rough Is Your Runway?
How Rough Is Your Runway? How Rough Is Your Runway?
How Rough Is Your Runway? Safe Software
 
Fast modelling of Airborne EM data using "Smart Interpretation"
Fast modelling of Airborne EM data using "Smart Interpretation"Fast modelling of Airborne EM data using "Smart Interpretation"
Fast modelling of Airborne EM data using "Smart Interpretation"Torben Bach
 
Prepare LiDAR Data To Meet Your Requirements
Prepare LiDAR Data To Meet Your RequirementsPrepare LiDAR Data To Meet Your Requirements
Prepare LiDAR Data To Meet Your RequirementsSafe Software
 
Stochastic kronecker graphs
Stochastic kronecker graphsStochastic kronecker graphs
Stochastic kronecker graphsZara Tariq
 
Colour Correction using Histogram Stretching
Colour Correction using Histogram StretchingColour Correction using Histogram Stretching
Colour Correction using Histogram StretchingPoul Kjeldager Sørensen
 
Modeling Count-based Raster Data with ArcGIS and R
Modeling Count-based Raster Data with ArcGIS and RModeling Count-based Raster Data with ArcGIS and R
Modeling Count-based Raster Data with ArcGIS and RAzavea
 

What's hot (20)

3D Analyst - Watershed, Padang
3D Analyst - Watershed, Padang3D Analyst - Watershed, Padang
3D Analyst - Watershed, Padang
 
3D Watershed Celebes
3D Watershed Celebes3D Watershed Celebes
3D Watershed Celebes
 
DSD-INT 2018 iMOD version 4.3 double precision big coordinates - Vermeulen
DSD-INT 2018 iMOD version 4.3 double precision big coordinates - VermeulenDSD-INT 2018 iMOD version 4.3 double precision big coordinates - Vermeulen
DSD-INT 2018 iMOD version 4.3 double precision big coordinates - Vermeulen
 
3D Analyst - Watershed, Lombok
3D Analyst - Watershed, Lombok3D Analyst - Watershed, Lombok
3D Analyst - Watershed, Lombok
 
Mutual information
Mutual informationMutual information
Mutual information
 
A Visualization Application for SIAT
A Visualization Application for SIATA Visualization Application for SIAT
A Visualization Application for SIAT
 
Internship
InternshipInternship
Internship
 
Undergraduate Modeling Workshop - Air Quality Working Group Final Presentatio...
Undergraduate Modeling Workshop - Air Quality Working Group Final Presentatio...Undergraduate Modeling Workshop - Air Quality Working Group Final Presentatio...
Undergraduate Modeling Workshop - Air Quality Working Group Final Presentatio...
 
Preparing LiDAR for Use in ArcGIS 10.1 with the Data Interoperability Extension
Preparing LiDAR for Use in ArcGIS 10.1 with the Data Interoperability ExtensionPreparing LiDAR for Use in ArcGIS 10.1 with the Data Interoperability Extension
Preparing LiDAR for Use in ArcGIS 10.1 with the Data Interoperability Extension
 
2017 Vendor Showcase Track: Tracking Z: Limitations of the World We Live In
2017 Vendor Showcase Track:  Tracking Z: Limitations of the World We Live In2017 Vendor Showcase Track:  Tracking Z: Limitations of the World We Live In
2017 Vendor Showcase Track: Tracking Z: Limitations of the World We Live In
 
Ecotect Presentaion
Ecotect PresentaionEcotect Presentaion
Ecotect Presentaion
 
3d hydrogeological conceptual model building in denmark
3d hydrogeological conceptual model building in denmark3d hydrogeological conceptual model building in denmark
3d hydrogeological conceptual model building in denmark
 
E Cognition User Summit2009 C Storch Gaf Emlc
E Cognition User Summit2009 C Storch Gaf EmlcE Cognition User Summit2009 C Storch Gaf Emlc
E Cognition User Summit2009 C Storch Gaf Emlc
 
Smart Interpretation - Fast AEM Modelling - SAGEEP 2017
Smart Interpretation - Fast AEM Modelling - SAGEEP 2017Smart Interpretation - Fast AEM Modelling - SAGEEP 2017
Smart Interpretation - Fast AEM Modelling - SAGEEP 2017
 
How Rough Is Your Runway?
How Rough Is Your Runway? How Rough Is Your Runway?
How Rough Is Your Runway?
 
Fast modelling of Airborne EM data using "Smart Interpretation"
Fast modelling of Airborne EM data using "Smart Interpretation"Fast modelling of Airborne EM data using "Smart Interpretation"
Fast modelling of Airborne EM data using "Smart Interpretation"
 
Prepare LiDAR Data To Meet Your Requirements
Prepare LiDAR Data To Meet Your RequirementsPrepare LiDAR Data To Meet Your Requirements
Prepare LiDAR Data To Meet Your Requirements
 
Stochastic kronecker graphs
Stochastic kronecker graphsStochastic kronecker graphs
Stochastic kronecker graphs
 
Colour Correction using Histogram Stretching
Colour Correction using Histogram StretchingColour Correction using Histogram Stretching
Colour Correction using Histogram Stretching
 
Modeling Count-based Raster Data with ArcGIS and R
Modeling Count-based Raster Data with ArcGIS and RModeling Count-based Raster Data with ArcGIS and R
Modeling Count-based Raster Data with ArcGIS and R
 

Similar to オープンハウスにおける 機械学習・データサイエンスの 取り組みについて

Stranger in a Srange Land;Exploring 3D and CityGML
Stranger in a Srange Land;Exploring 3D and CityGMLStranger in a Srange Land;Exploring 3D and CityGML
Stranger in a Srange Land;Exploring 3D and CityGMLSafe Software
 
How Data Science can help energy companies map their infrastructure
How Data Science can help energy companies map their infrastructureHow Data Science can help energy companies map their infrastructure
How Data Science can help energy companies map their infrastructureAlex Combessie
 
Introducing google’s mobile nets
Introducing google’s mobile netsIntroducing google’s mobile nets
Introducing google’s mobile netsLarry Guo
 
CityGML Integration Into the ArcGIS Platform
CityGML Integration Into the ArcGIS PlatformCityGML Integration Into the ArcGIS Platform
CityGML Integration Into the ArcGIS PlatformSafe Software
 
Field Geometry, auto steering and services
Field Geometry, auto steering and servicesField Geometry, auto steering and services
Field Geometry, auto steering and servicesCAPIGI
 
Analysis of data science software 2020
Analysis of data science software 2020Analysis of data science software 2020
Analysis of data science software 2020Russ Reinsch
 
Smart Urban Planning Support through Web Data Science on Open and Enterprise ...
Smart Urban Planning Support through Web Data Science on Open and Enterprise ...Smart Urban Planning Support through Web Data Science on Open and Enterprise ...
Smart Urban Planning Support through Web Data Science on Open and Enterprise ...Gloria Re Calegari
 
T3.2 Application domain extention (ADE)
T3.2 Application domain extention (ADE)T3.2 Application domain extention (ADE)
T3.2 Application domain extention (ADE)i-SCOPE Project
 
2017 PLSC Track: Using a Standard Version of ArcMap with External VRS Recieve...
2017 PLSC Track: Using a Standard Version of ArcMap with External VRS Recieve...2017 PLSC Track: Using a Standard Version of ArcMap with External VRS Recieve...
2017 PLSC Track: Using a Standard Version of ArcMap with External VRS Recieve...GIS in the Rockies
 
2018 - Grupo QGIS Brasil e o lançamento do QGIS 3.4 LTR (Versão de Longo Prazo)
2018 - Grupo QGIS Brasil e o lançamento do QGIS 3.4 LTR (Versão de Longo Prazo)2018 - Grupo QGIS Brasil e o lançamento do QGIS 3.4 LTR (Versão de Longo Prazo)
2018 - Grupo QGIS Brasil e o lançamento do QGIS 3.4 LTR (Versão de Longo Prazo)George Porto Ferreira
 
IoT Workload Distribution Impact Between Edge and Cloud Computing in a Smart ...
IoT Workload Distribution Impact Between Edge and Cloud Computing in a Smart ...IoT Workload Distribution Impact Between Edge and Cloud Computing in a Smart ...
IoT Workload Distribution Impact Between Edge and Cloud Computing in a Smart ...Otávio Carvalho
 
“COVID-19 Safe Distancing Measures in Public Spaces with Edge AI,” a Presenta...
“COVID-19 Safe Distancing Measures in Public Spaces with Edge AI,” a Presenta...“COVID-19 Safe Distancing Measures in Public Spaces with Edge AI,” a Presenta...
“COVID-19 Safe Distancing Measures in Public Spaces with Edge AI,” a Presenta...Edge AI and Vision Alliance
 
Geotagging Social Media Content with a Refined Language Modelling Approach
Geotagging Social Media Content with a Refined Language Modelling ApproachGeotagging Social Media Content with a Refined Language Modelling Approach
Geotagging Social Media Content with a Refined Language Modelling ApproachSymeon Papadopoulos
 
Geotagging Social Media Content with a Refined Language Modelling Approach
Geotagging Social Media Content with a Refined Language Modelling ApproachGeotagging Social Media Content with a Refined Language Modelling Approach
Geotagging Social Media Content with a Refined Language Modelling ApproachREVEAL - Social Media Verification
 
Introduction to mago3D, an Open Source Based Digital Twin Platform
Introduction to mago3D, an Open Source Based Digital Twin PlatformIntroduction to mago3D, an Open Source Based Digital Twin Platform
Introduction to mago3D, an Open Source Based Digital Twin PlatformSANGHEE SHIN
 
Minimum image disortion of reversible data hiding
Minimum image disortion of reversible data hidingMinimum image disortion of reversible data hiding
Minimum image disortion of reversible data hidingIRJET Journal
 
Visicom 3d models usage
Visicom 3d models usageVisicom 3d models usage
Visicom 3d models usageSaboor Marwat
 
IRJET- 3D Object Recognition of Car Image Detection
IRJET-  	  3D Object Recognition of Car Image DetectionIRJET-  	  3D Object Recognition of Car Image Detection
IRJET- 3D Object Recognition of Car Image DetectionIRJET Journal
 
How to Automate CAD & GIS Integration
How to Automate CAD & GIS IntegrationHow to Automate CAD & GIS Integration
How to Automate CAD & GIS IntegrationSafe Software
 

Similar to オープンハウスにおける 機械学習・データサイエンスの 取り組みについて (20)

Stranger in a Srange Land;Exploring 3D and CityGML
Stranger in a Srange Land;Exploring 3D and CityGMLStranger in a Srange Land;Exploring 3D and CityGML
Stranger in a Srange Land;Exploring 3D and CityGML
 
How Data Science can help energy companies map their infrastructure
How Data Science can help energy companies map their infrastructureHow Data Science can help energy companies map their infrastructure
How Data Science can help energy companies map their infrastructure
 
Introducing google’s mobile nets
Introducing google’s mobile netsIntroducing google’s mobile nets
Introducing google’s mobile nets
 
CityGML Integration Into the ArcGIS Platform
CityGML Integration Into the ArcGIS PlatformCityGML Integration Into the ArcGIS Platform
CityGML Integration Into the ArcGIS Platform
 
Masters Thesis
Masters ThesisMasters Thesis
Masters Thesis
 
Field Geometry, auto steering and services
Field Geometry, auto steering and servicesField Geometry, auto steering and services
Field Geometry, auto steering and services
 
Analysis of data science software 2020
Analysis of data science software 2020Analysis of data science software 2020
Analysis of data science software 2020
 
Smart Urban Planning Support through Web Data Science on Open and Enterprise ...
Smart Urban Planning Support through Web Data Science on Open and Enterprise ...Smart Urban Planning Support through Web Data Science on Open and Enterprise ...
Smart Urban Planning Support through Web Data Science on Open and Enterprise ...
 
T3.2 Application domain extention (ADE)
T3.2 Application domain extention (ADE)T3.2 Application domain extention (ADE)
T3.2 Application domain extention (ADE)
 
2017 PLSC Track: Using a Standard Version of ArcMap with External VRS Recieve...
2017 PLSC Track: Using a Standard Version of ArcMap with External VRS Recieve...2017 PLSC Track: Using a Standard Version of ArcMap with External VRS Recieve...
2017 PLSC Track: Using a Standard Version of ArcMap with External VRS Recieve...
 
2018 - Grupo QGIS Brasil e o lançamento do QGIS 3.4 LTR (Versão de Longo Prazo)
2018 - Grupo QGIS Brasil e o lançamento do QGIS 3.4 LTR (Versão de Longo Prazo)2018 - Grupo QGIS Brasil e o lançamento do QGIS 3.4 LTR (Versão de Longo Prazo)
2018 - Grupo QGIS Brasil e o lançamento do QGIS 3.4 LTR (Versão de Longo Prazo)
 
IoT Workload Distribution Impact Between Edge and Cloud Computing in a Smart ...
IoT Workload Distribution Impact Between Edge and Cloud Computing in a Smart ...IoT Workload Distribution Impact Between Edge and Cloud Computing in a Smart ...
IoT Workload Distribution Impact Between Edge and Cloud Computing in a Smart ...
 
“COVID-19 Safe Distancing Measures in Public Spaces with Edge AI,” a Presenta...
“COVID-19 Safe Distancing Measures in Public Spaces with Edge AI,” a Presenta...“COVID-19 Safe Distancing Measures in Public Spaces with Edge AI,” a Presenta...
“COVID-19 Safe Distancing Measures in Public Spaces with Edge AI,” a Presenta...
 
Geotagging Social Media Content with a Refined Language Modelling Approach
Geotagging Social Media Content with a Refined Language Modelling ApproachGeotagging Social Media Content with a Refined Language Modelling Approach
Geotagging Social Media Content with a Refined Language Modelling Approach
 
Geotagging Social Media Content with a Refined Language Modelling Approach
Geotagging Social Media Content with a Refined Language Modelling ApproachGeotagging Social Media Content with a Refined Language Modelling Approach
Geotagging Social Media Content with a Refined Language Modelling Approach
 
Introduction to mago3D, an Open Source Based Digital Twin Platform
Introduction to mago3D, an Open Source Based Digital Twin PlatformIntroduction to mago3D, an Open Source Based Digital Twin Platform
Introduction to mago3D, an Open Source Based Digital Twin Platform
 
Minimum image disortion of reversible data hiding
Minimum image disortion of reversible data hidingMinimum image disortion of reversible data hiding
Minimum image disortion of reversible data hiding
 
Visicom 3d models usage
Visicom 3d models usageVisicom 3d models usage
Visicom 3d models usage
 
IRJET- 3D Object Recognition of Car Image Detection
IRJET-  	  3D Object Recognition of Car Image DetectionIRJET-  	  3D Object Recognition of Car Image Detection
IRJET- 3D Object Recognition of Car Image Detection
 
How to Automate CAD & GIS Integration
How to Automate CAD & GIS IntegrationHow to Automate CAD & GIS Integration
How to Automate CAD & GIS Integration
 

More from Teito Nakagawa

Object Detection on AWS Lambda
Object Detection on AWS LambdaObject Detection on AWS Lambda
Object Detection on AWS LambdaTeito Nakagawa
 
BigQuery GISを用いた物件レコメンド
BigQuery GISを用いた物件レコメンドBigQuery GISを用いた物件レコメンド
BigQuery GISを用いた物件レコメンドTeito Nakagawa
 
Numacraw for r user(upload)
Numacraw for r user(upload)Numacraw for r user(upload)
Numacraw for r user(upload)Teito Nakagawa
 
Numacraw for r user(upload)
Numacraw for r user(upload)Numacraw for r user(upload)
Numacraw for r user(upload)Teito Nakagawa
 
Stanで人類最強の男を決定する 2
Stanで人類最強の男を決定する 2Stanで人類最強の男を決定する 2
Stanで人類最強の男を決定する 2Teito Nakagawa
 
Collaborativefilteringwith r
Collaborativefilteringwith rCollaborativefilteringwith r
Collaborativefilteringwith rTeito Nakagawa
 

More from Teito Nakagawa (8)

Object Detection on AWS Lambda
Object Detection on AWS LambdaObject Detection on AWS Lambda
Object Detection on AWS Lambda
 
BigQuery GISを用いた物件レコメンド
BigQuery GISを用いた物件レコメンドBigQuery GISを用いた物件レコメンド
BigQuery GISを用いた物件レコメンド
 
Numacraw for r user(upload)
Numacraw for r user(upload)Numacraw for r user(upload)
Numacraw for r user(upload)
 
Numacraw for r user(upload)
Numacraw for r user(upload)Numacraw for r user(upload)
Numacraw for r user(upload)
 
Stanで人類最強の男を決定する 2
Stanで人類最強の男を決定する 2Stanで人類最強の男を決定する 2
Stanで人類最強の男を決定する 2
 
StanTutorial
StanTutorialStanTutorial
StanTutorial
 
Introduction of stan
Introduction of stanIntroduction of stan
Introduction of stan
 
Collaborativefilteringwith r
Collaborativefilteringwith rCollaborativefilteringwith r
Collaborativefilteringwith r
 

Recently uploaded

Pigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping ElbowsPigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping ElbowsPigging Solutions
 
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...shyamraj55
 
costume and set research powerpoint presentation
costume and set research powerpoint presentationcostume and set research powerpoint presentation
costume and set research powerpoint presentationphoebematthew05
 
Unraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdfUnraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdfAlex Barbosa Coqueiro
 
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationBeyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationSafe Software
 
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmaticsKotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmaticscarlostorres15106
 
Transcript: New from BookNet Canada for 2024: BNC BiblioShare - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC BiblioShare - Tech Forum 2024Transcript: New from BookNet Canada for 2024: BNC BiblioShare - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC BiblioShare - Tech Forum 2024BookNet Canada
 
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...Patryk Bandurski
 
AI as an Interface for Commercial Buildings
AI as an Interface for Commercial BuildingsAI as an Interface for Commercial Buildings
AI as an Interface for Commercial BuildingsMemoori
 
SIP trunking in Janus @ Kamailio World 2024
SIP trunking in Janus @ Kamailio World 2024SIP trunking in Janus @ Kamailio World 2024
SIP trunking in Janus @ Kamailio World 2024Lorenzo Miniero
 
Vertex AI Gemini Prompt Engineering Tips
Vertex AI Gemini Prompt Engineering TipsVertex AI Gemini Prompt Engineering Tips
Vertex AI Gemini Prompt Engineering TipsMiki Katsuragi
 
Install Stable Diffusion in windows machine
Install Stable Diffusion in windows machineInstall Stable Diffusion in windows machine
Install Stable Diffusion in windows machinePadma Pradeep
 
"ML in Production",Oleksandr Bagan
"ML in Production",Oleksandr Bagan"ML in Production",Oleksandr Bagan
"ML in Production",Oleksandr BaganFwdays
 
APIForce Zurich 5 April Automation LPDG
APIForce Zurich 5 April  Automation LPDGAPIForce Zurich 5 April  Automation LPDG
APIForce Zurich 5 April Automation LPDGMarianaLemus7
 
Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024Scott Keck-Warren
 
Dev Dives: Streamline document processing with UiPath Studio Web
Dev Dives: Streamline document processing with UiPath Studio WebDev Dives: Streamline document processing with UiPath Studio Web
Dev Dives: Streamline document processing with UiPath Studio WebUiPathCommunity
 
Nell’iperspazio con Rocket: il Framework Web di Rust!
Nell’iperspazio con Rocket: il Framework Web di Rust!Nell’iperspazio con Rocket: il Framework Web di Rust!
Nell’iperspazio con Rocket: il Framework Web di Rust!Commit University
 
Designing IA for AI - Information Architecture Conference 2024
Designing IA for AI - Information Architecture Conference 2024Designing IA for AI - Information Architecture Conference 2024
Designing IA for AI - Information Architecture Conference 2024Enterprise Knowledge
 
Unblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen FramesUnblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen FramesSinan KOZAK
 

Recently uploaded (20)

Pigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping ElbowsPigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping Elbows
 
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
 
costume and set research powerpoint presentation
costume and set research powerpoint presentationcostume and set research powerpoint presentation
costume and set research powerpoint presentation
 
Unraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdfUnraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdf
 
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationBeyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
 
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmaticsKotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
 
Transcript: New from BookNet Canada for 2024: BNC BiblioShare - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC BiblioShare - Tech Forum 2024Transcript: New from BookNet Canada for 2024: BNC BiblioShare - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC BiblioShare - Tech Forum 2024
 
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
 
AI as an Interface for Commercial Buildings
AI as an Interface for Commercial BuildingsAI as an Interface for Commercial Buildings
AI as an Interface for Commercial Buildings
 
SIP trunking in Janus @ Kamailio World 2024
SIP trunking in Janus @ Kamailio World 2024SIP trunking in Janus @ Kamailio World 2024
SIP trunking in Janus @ Kamailio World 2024
 
Vertex AI Gemini Prompt Engineering Tips
Vertex AI Gemini Prompt Engineering TipsVertex AI Gemini Prompt Engineering Tips
Vertex AI Gemini Prompt Engineering Tips
 
Install Stable Diffusion in windows machine
Install Stable Diffusion in windows machineInstall Stable Diffusion in windows machine
Install Stable Diffusion in windows machine
 
"ML in Production",Oleksandr Bagan
"ML in Production",Oleksandr Bagan"ML in Production",Oleksandr Bagan
"ML in Production",Oleksandr Bagan
 
APIForce Zurich 5 April Automation LPDG
APIForce Zurich 5 April  Automation LPDGAPIForce Zurich 5 April  Automation LPDG
APIForce Zurich 5 April Automation LPDG
 
Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024
 
Dev Dives: Streamline document processing with UiPath Studio Web
Dev Dives: Streamline document processing with UiPath Studio WebDev Dives: Streamline document processing with UiPath Studio Web
Dev Dives: Streamline document processing with UiPath Studio Web
 
Nell’iperspazio con Rocket: il Framework Web di Rust!
Nell’iperspazio con Rocket: il Framework Web di Rust!Nell’iperspazio con Rocket: il Framework Web di Rust!
Nell’iperspazio con Rocket: il Framework Web di Rust!
 
Designing IA for AI - Information Architecture Conference 2024
Designing IA for AI - Information Architecture Conference 2024Designing IA for AI - Information Architecture Conference 2024
Designing IA for AI - Information Architecture Conference 2024
 
Vulnerability_Management_GRC_by Sohang Sengupta.pptx
Vulnerability_Management_GRC_by Sohang Sengupta.pptxVulnerability_Management_GRC_by Sohang Sengupta.pptx
Vulnerability_Management_GRC_by Sohang Sengupta.pptx
 
Unblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen FramesUnblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen Frames
 

オープンハウスにおける 機械学習・データサイエンスの 取り組みについて

  • 4.
  • 5.
  • 6.
  • 7. IT IT Dev OpsData • Smartphone App • Agile Development • Device • Cloud • Help Desk • Data Science • Analytics *http://www.itmedia.co.jp/enterprise/articles/ 1802/26/news007.html *http://diamond.jp/articles/-/150122
  • 9. 20182017 CRM/SFA MA D-Marketing IoT Collab. Phase 3 Insighti Real Estate SPA DD-Biz. ML/DL Big Data PoC Phase 2 Agility (2/2)
  • 11.
  • 14. ( F LVZ • KPSJQ C8B G • 1 G 1 G 1 • O N , C 4 • C8B G ) • aW I • JTI CG 7 BigQuery
  • 15.
  • 18.
  • 19.
  • 20. • or • or or Web …
  • 21. Q. A. – – Q. or A. – – A B C D 3:7 2:2 →
  • 23.
  • 25. ) ( 0
  • 26. Comparison of urban land price prediction approaches
  • 27. Data : introduction ● Public Land Price data released by the Ministry of Land, Infrastructure, Transport and Tourism as of 2018/01/01 ● Area Tokyo, Nagoya, Kanagawa, Chiba and Saitama (without Tokyo Islands) ● Type of lands Residential
  • 28. Model 1 : Geographically Weighted Regression (GWR) “Everything is related to everything else. But near things are more related than distant things” Tobler’s first law of Geography Normal Regression Every point is treated the same for prediction GWR Closer points are treated as more important : the closer the bigger the weight 10 30 30 10 20 10 30 30 10 20 Prediction: 20 Prediction: 12
  • 29. ● Model definition: with yi the land price, xik the value of the variable k, (ui,vi) the coordinates, βk the regression parameter for the variable k and εi the error at location i ● Parameter estimation (regression) in matrix notation: with W(ui,vi) the diagonal matrix denoting the geographical weighting of each observed data for regression point i Model 1 : Geographically Weighted Regression (GWR)
  • 30. ● Weight at regression point i of datapoint j : ● Bandwidth (scale) selection with a Golden-Section algorithm ● Bandwidth is adaptive rather than fixed : Adaptive bandwidth that includes the k-nearest neighbors at each regression point rather than a fixed value This allows to incorporate data points with few close neighbors better in the regression Model 1 : Geographically Weighted Regression (GWR) where dij is the distance between i and j, and b is the bandwidth or scale
  • 31. Model 2 : Hedonic ● Model definition : ● Regression method: OLS Yi Land Price in /m2 (log) Variables (xk) Units Explanations Distance to closest station m (log) Distance to big railway hubs m (log) ( : : ...) Floor-area ratio % Road width m Gas Flag : connected to gas infrastructure or not Land area m² (log) Building material Categorical concrete, wood,... Land usage Categorical low-rise residential type-1, semi-residential, ... Neighborhood weighted prices distance-weighted average of the 9 nearest neighbors
  • 32. Model 3 : Boosted Trees (XGBoost) ● A set of weak learners (decision trees) are combined to get strong learners ● Trees are grown sequentially : each tree is grown using information from previously grown trees ● Boosted trees are implemented using XGBoost library ● Same variables as Model 2 (Hedonic) Distance to closest station, Distance to big railway hubs, Floor-area ratio, Road width, Gas, Land area, Building material, Land usage, Neighborhood weighted prices ERRORS TRAIN TEST TRAIN TEST MODEL MODEL DATASET
  • 33. Results : Ratios of low-error predictions Error Global Tokyo Saitama Kanagawa Chiba Nagoya GWR < 5 % 32 % 39 % 27 % 34 % 22 % 34 % < 10% 57 % 63 % 57 % 62 % 37 % 59 % < 20% 83 % 86 % 85 % 89 % 67 % 87 % Hedonic < 5 % 34 % 38 % 32 % 33 % 21 % 34% < 10% 56 % 63 % 64 % 57 % 41 % 59 % < 20% 83 % 85 % 89 % 85 % 64 % 87 % XGBoost < 5 % 35 % 41 % 34 % 35 % 21 % 35 % < 10% 61 % 66 % 66 % 63 % 42 % 61 % < 20% 85 % 92 % 88 % 88 % 69 % 88 %
  • 34. Results : prediction error distribution GWR 11.9 % 8.4 % Hedonic 11.9 % 8.7 % XGBoost 11.3 % 8.3 % ● Error estimation via a 100-fold Cross Validation (train / eval to data ratio : 75% / 25%) Mean prediction error by area Error distribution for each model
  • 35. Results : maps Predicted prices map around Tokyo Prediction error around Tokyo
  • 36. Results : discussion XGBoost was the best performing model ● Model Limitations and possible ameliorations ○ GWR The current version is single scale, assuming that all the variables experience local effects on the same scale. Multiscale GWR would drop that assumption and potentially improve the accuracy ○ Hedonic/XGBoost : Spatial correlation effects are treated empirically ● Data : ○ Data points : some areas have little data point, dragging the overall accuracy down (especially Chiba) ○ Euclidean distance is used for in each case Other distance metrics (Manhattan distance, commute time,..) might be more suited
  • 37. • IBIS 2018 2018 11 • /