Muutke küpsiste eelistusi

E-raamat: Public Policy Analytics: Code and Context for Data Science in Government [Taylor & Francis e-raamat]

  • Formaat: 228 pages, 22 Tables, black and white; 116 Line drawings, color; 116 Illustrations, color
  • Sari: Chapman & Hall/CRC Data Science Series
  • Ilmumisaeg: 20-Aug-2021
  • Kirjastus: CRC Press
  • ISBN-13: 9781003054658
Teised raamatud teemal:
  • Taylor & Francis e-raamat
  • Hind: 207,73 €*
  • * hind, mis tagab piiramatu üheaegsete kasutajate arvuga ligipääsu piiramatuks ajaks
  • Tavahind: 296,75 €
  • Säästad 30%
Public Policy Analytics: Code and Context for Data Science in GovernmentSuurem pilt
  • Formaat: 228 pages, 22 Tables, black and white; 116 Line drawings, color; 116 Illustrations, color
  • Sari: Chapman & Hall/CRC Data Science Series
  • Ilmumisaeg: 20-Aug-2021
  • Kirjastus: CRC Press
  • ISBN-13: 9781003054658
Teised raamatud teemal:
Taylor & Francis e-raamatudRohkem infot Taylor & Francis e-raamatute kohta
Raamatu kodulehekülg: https://www.taylorfrancis.com/books/9781003054658
"Public Policy Analytics: Code & Context for Data Science in Government teaches readers how to address complex public policy problems with data and analytics using reproducible methods in R. Each of the eight chapters provides a detailed case study, showing readers: how to develop exploratory indicators; understand 'spatial process' and develop spatial analytics; how to develop 'useful' predictive analytics; how to convey these outputs to non-technical decision-makers through the medium of data visualization; and why, ultimately, data science and 'Planning' are one and the same. A graduate-level introduction to data science, this book will appeal to researchers and data scientists at the intersection of data analytics and public policy, as well as readers who wish to understand how algorithms will affect the future of government"--

Public Policy Analytics: Code & Context for Data Science in Government teaches readers how to address complex public policy problems with data and analytics using reproducible methods in R. Each of the eight chapters provides a detailed case study, showing readers: how to develop exploratory indicators; understand ‘spatial process’ and develop spatial analytics; how to develop ‘useful’ predictive analytics; how to convey these outputs to non-technical decision-makers through the medium of data visualization; and why, ultimately, data science and ‘Planning’ are one and the same. A graduate-level introduction to data science, this book will appeal to researchers and data scientists at the intersection of data analytics and public policy, as well as readers who wish to understand how algorithms will affect the future of government.

About the Author ix
Preface xi
Introduction xiii
How governments make decisions
xiii
Context as the foundation
xv
Data science as a planning tool
xvi
The importance of spatial thinking
xvii
Learning objectives
xx
1 Indicators for Transit-oriented Development 1(22)
1.1 Why start with indicators?
1(4)
1.1.1 Mapping and scale bias in areal aggregate data
2(3)
1.2 Set up
5(10)
1.2.1 Downloading and wrangling census data
6(5)
1.2.2 Wrangling transit open data
11(1)
1.2.3 Relating tracts and subway stops in space
12(3)
1.3 Developing TOD indicators
15(3)
1.3.1 TOD indicator maps
15(1)
1.3.2 TOD indicator tables
16(1)
1.3.3 TOD indicator plots
17(1)
1.4 Capturing three submarkets of interest
18(2)
1.5 Conclusion: Are Philadelphians willing to pay for TOD?
20(1)
1.6 Assignment - Study TOD in your city
21(2)
2 Expanding the Urban Growth Boundary 23(22)
2.1 Introduction - Lancaster development
23(7)
2.1.1 The bid-rent model
26(3)
2.1.2 Set up Lancaster data
29(1)
2.2 Identifying areas inside and outside of the Urban Growth Area
30(7)
2.2.1 Associate each inside/outside buffer with its respective town
32(1)
2.2.2 Building density by town and by inside/outside the UGA
33(1)
2.2.3 Visualize buildings inside and outside the UGA
34(3)
2.3 Return to Lancaster's bid-rent
37(4)
2.4 Conclusion - On boundaries
41(1)
2.5 Assignment - Boundaries in your community
41(4)
3 Intro to Geospatial Machine Learning, Part 1 45(26)
3.1 Machine learning as a planning
45(3)
3.1.1 Accuracy and generalizability
46(1)
3.1.2 The machine learning process
46(1)
3.1.3 The hedonic model
47(1)
3.2 Data wrangling - Home price and crime data
48(8)
3.2.1 Feature engineering - Measuring exposure to crime
51(2)
3.2.2 Exploratory analysis - Correlation
53(3)
3.3 Introduction to ordinary least squares regression
56(7)
3.3.1 Our first regression model
58(2)
3.3.2 More feature engineering and colinearity
60(3)
3.4 Cross-validation and return to goodness of fit
63(5)
3.4.1 Accuracy - Mean absolute error
63(2)
3.4.2 Generalizability - Cross-validation
65(3)
3.5 Conclusion - Our first model
68(1)
3.6 Assignment - Predict house prices
68(3)
4 Intro to Geospatial Machine Learning, Part 2 71(16)
4.1 On the spatial process of home prices
71(3)
4.1.1 Set up and data wrangling
72(2)
4.2 Do prices and errors cluster? The spatial lag
74(3)
4.2.1 Do model errors cluster? - Moran's I
75(2)
4.3 Accounting for neighborhood
77(8)
4.3.1 Accuracy of the neighborhood model
78(2)
4.3.2 Spatial autocorrelation in the neighborhood model
80(2)
4.3.3 Generalizability of the neighborhood model
82(3)
4.4 Conclusion - Features at multiple scales
85(2)
5 Geospatial Risk Modeling - Predictive Policing 87(42)
5.1 New predictive policing tools
87(4)
5.1.1 Generalizability in geospatial risk models
88(1)
5.1.2 From broken windows theory to broken windows policing
89(1)
5.1.3 Set up
90(1)
5.2 Data wrangling: Creating the fishnet
91(7)
5.2.1 Data wrangling: Joining burglaries to the fishnet
94(1)
5.2.2 Wrangling risk factors
95(3)
5.3 Feature engineering - Count of risk factors by grid cell
98(6)
5.3.1 Feature engineering - Nearest neighbor features
100(2)
5.3.2 Feature Engineering - Measure distance to one point
102(1)
5.3.3 Feature Engineering - Create the final_net
103(1)
5.4 Exploring the spatial process of burglary
104(5)
5.4.1 Correlation tests
108(1)
5.5 Poisson Regression
109(16)
5.5.1 Cross-validated Poisson regression
111(1)
5.5.2 Accuracy and generalzability
112(6)
5.5.3 Generalizability by neighborhood context
118(2)
5.5.4 Does this model allocate better than traditional crime hotspots?
120(5)
5.6 Conclusion - Bias but useful?
125(2)
5.7 Assignment - Predict risk
127(2)
6 People-based ML Models 129(24)
6.1 Bounce to work
129(2)
6.2 Exploratory analysis
131(3)
6.3 Logistic regression
134(3)
6.3.1 Training/testing sets
135(1)
6.3.2 Estimate a churn model
135(2)
6.4 Goodness of fit
137(5)
6.4.1 Roc curves
141(1)
6.5 Cross-validation
142(2)
6.6 Generating costs and benefits
144(5)
6.6.1 Optimizing the cost/benefit relationship
146(3)
6.7 Conclusion - Churn
149(1)
6.8 Assignment - Target a subsidy
150(3)
7 People-based ML Models: Algorithmic Fairness 153(18)
7.1 Introduction
153(3)
7.1.1 The specter of disparate impact
154(1)
7.1.2 Modeling judicial outcomes
155(1)
7.1.3 Accuracy and generalizability in recidivism algorithms
155(1)
7.2 Data and exploratory analysis
156(3)
7.3 Estimate two recidivism models
159(5)
7.3.1 Accuracy and generalizability
161(3)
7.4 What about the threshold?
164(1)
7.5 Optimizing 'equitable' thresholds
165(3)
7.6 Assignment - Memo to the mayor
168(3)
8 Predicting Rideshare Demand 171(26)
8.1 Introduction - Rideshare
171(1)
8.2 Data wrangling - Rideshare
172(7)
8.2.1 Lubridate
173(1)
8.2.2 Weather data
174(1)
8.2.3 Subset a study area using neighborhoods
175(2)
8.2.4 Create the final space/time panel
177(1)
8.2.5 Split training and test
178(1)
8.2.6 What about distance features?
179(1)
8.3 Exploratory Analysis - Rideshare
179(8)
8.3.1 Trip_Count serial autocorrelation
180(2)
8.3.2 Trip_Count spatial autocorrelation
182(2)
8.3.3 Space/time correlation?
184(1)
8.3.4 Weather
185(2)
8.4 Modeling and validation using purrr ::map
187(7)
8.4.1 A short primer on nested tibbles
187(1)
8.4.2 Estimate a rideshare forecast
188(1)
8.4.3 Validate test set by time
189(3)
8.4.4 Validate test set by space
192(2)
8.5 Conclusion - Dispatch
194(1)
8.6 Assignment - Predict bike share trips
195(2)
Conclusion - Algorithmic Governance 197(4)
Index 201
Ken Steif Ph.D is the Director of the Master of Urban Spatial Analytics program at the University of Pennsylvania and an Associate Professor of Practice in the City Planning Program. He teaches courses on the application of spatial analysis, statistics, predictive modeling and data visualization to solve traditional and contemporary public policy programs. Dr. Steif also is the founder of a consultancy that develops analytics for both public and private-sector clients. He lives in West Philadelphia with his wife Diana and sons Emil and Malcolm. You can follow him on Twitter @KenSteif.
Püsilink: https://www.kriso.ee/db/9781003054658_pe.html
Märksõnad: