Geospatial Analytics Use cases - Assignment

RUNNING HEAD: Applications of GIS data and spatial thinking role in data science
Title: Application of GIS data and spatial thinking role in data science
Student Name:
Student Institution:

Running Head: Application of GIS data and spatial thinking role in data science 2
Abstract
This report examines the strengths and weaknesses of currently available off-the-shelf
versions of GIS as a learning environment. In making its judgment on the design and
implementation of GIS as a support system for spatial thinking in the data science environment,
the committee relies on primarily oral presentations and written statements from system
designers, researchers, and school and university educators.
Introduction
Geographers use GIS (Geographic Information System) to document, study, and
communicate information about a wide array of things we find all around us every day. In this
lab we will explore how GIS is used to study questions of environmental justice. GIS are used by
businesses (Google, Microsoft, Amazon, UPS, CVS, and on and on), local government agencies
(City of Omaha, Omaha Metro, OPPD, Omaha MUD, etc.), federal agencies (US Department of
Agriculture, US Forest Service, US Geological Survey, US Fish and Wildlife Service, Census
Bureau, etc.), and scientists and academics doing social and environmental research. Developing
techniques, tools, and expert application of GIS for the groups listed above are specializations
within the field of Geography.
Information analysis and Data synthesis
GIS are powerful tools for collecting, storing, organizing, doing statistical analysis, doing
spatial analysis, reporting, visualizing, and mapping spatial data. Spatial data are any data that
can be associated with specific places – locations. For instance home prices in a neighborhood –
with GIS not only can you make a map of where homes are, but you can do an analysis of how
home prices vary with distances from schools, parks, or restaurants (Agrawal et al. 1993).

Running Head: Application of GIS data and spatial thinking role in data science 3
Two commonly used analogies for GIS are the “shoe box” (or recipe box) and the “tool
box” analogies. Thinking of GIS as a shoe box indicates the GIS is used as a way to store and
organize information. You may know someone who has a collection of recipes on cards (either
on paper or in a computer application). In the shoe box analogy the cards are stored and
organized systematically in the shoe box. A geographic example of this can be found in your
neighborhood. Think of the objects found outside around the place where you live – the streets,
the storm drains, the fire hydrants, the trees, the houses. In the shoe box analogy each item, every
street, every storm drain, etc. gets its own card with a description of it, and gets stored in the
shoe box (the GIS database). An important distinction for GIS from a simple shoe box is that
each card has spatial information – map coordinates so the objects can be placed in geographic
relation to each other in 2 or 3 dimensions. The cards are not just put in the shoe box in any order
– all of the streets are together in a group of their own called a layer, the fire hydrants are all
together in a fire hydrant layer, etc. This organization allows us to map any type of feature layer
individually or in combination with other layers, and search and analyze the specific information
found on the individual cards and displays it in a map (Agrawal et al. 1993).
The tool box analogy is a way of seeing GIS as a set of tools used to work on spatial data.
Like a real tool box there are many tools in a GIS and each tool has a specific purpose used for
different tasks or for different types of data. The tools available in a GIS include tools for visual
display – mapping tools – tools for statistical analysis – based measurements related to the
objects in the shoe box – tools for spatial analysis – based on the geographic location of objects
in the map or relative to each other (distance, size, shape, direction, etc.) – and tools for data
conversion, projecting, selecting, copying, and joining data sets together. Having all of these

Running Head: Application of GIS data and spatial thinking role in data science 4
tools together to work with spatial data is a project begun by geographers in the 1960s and
continues today (Anthony et al. 2007, Duckham & Kulik 2006).
One application of GIS to a human-environment problem is environmental justice.
Environmental justice is concerned with who is affected disproportionately by poor or hazardous
environmental quality. Environmental justice is a concern because over time it has become
empirically evident that some socioeconomic groups tend to be located in or near areas of
degraded environmental quality or potential environmental hazards, putting them at greater risk
for damaged health or economic loss (Anthony et al. 2007, Duckham & Kulik 2006).
In a society that uses and degrades the environment creating both goods and waste,
someone will benefit from and someone will pay the costs of this use (harvesting natural
resources, consuming water, mining, etc.) and degradation (pollution generated in the process,
depletion of resources).A major factor that influences environmental justice is the distribution of
power in society – social, economic, and political power – the power to make decisions and
influence one’s own environment and the environment of others. This why groups with less
socioeconomic and political power often lives or works in poor or hazardous environmental
conditions. Typically we see racial minorities, working class, immigrant groups, and women,
working and living in poor or hazardous conditions. On the other hand, the racial majority, the
wealthy, and the socially and politically connected have better living and working conditions –
they have the power – or have ingratiated themselves to power – to protect themselves from poor
or hazardous environments. Note that this is true in countries around the world, not just in the
United States (Anthony et al. 2007, Duckham & Kulik 2006).

Running Head: Application of GIS data and spatial thinking role in data science 5
The ability of GIS to organize, visualize, and analyze data about where different
socioeconomic groups and environmental hazards are located makes it especially useful for
studying and communicating problems of environmental justice.
While one can analyze the distribution or segregation of minority residents in a city, and
count the number of oil refineries (for example); in a GIS one can combine this information with
location to understand if the two things are found in the same place, how often, and if other
groups are affected in the same way.
In this report we will use a GIS of Douglas County to visualize and analyze data from the
US Census and the Environmental Protection Agency’s (EPA) Toxics Release Inventory (TRI).
Both the US Census data and EPA data are frequently used by geographers interested in
demographic, environmental, and human-environment questions (Anthony et al. 2007, Duckham
& Kulik 2006).
The US Census is conducted every ten years to understand the characteristics and
changes in the United States Population. The census data are publicly available and frequently
used by geographers to understand and explore social, cultural, and economic aspects of the

Running Head: Application of GIS data and spatial thinking role in data science 6
United States. Census data are particularly geographic since they are grouped or aggregated
(individual responses to census questions are summarized by geographic area).
The EPA TRI program is intended to track the use of toxic chemicals that are potentially
harmful to human health and the environment. The TRI program requires that facilities report
how much each monitored chemical is released to the environment each year, including
chemicals that are recycled or treated after use (Anthony et al. 2007, Duckham & Kulik 2006).
It depends on an ordinary decoration of square cells, yet exploits situations where
neighboring cells have a similar field esteem, with the goal that they can together be spoken to as
one greater cell. A straightforward delineation is given in the figure beneath. It shows a little 8 Ã
—8 raster with three potential field esteems: white, green and blue. The quad tree that speaks to
this raster is developed by more than once separating the zone into four quadrants, which are
called NW, NE, SE, and SW for clear reasons. This method stops when every one of the cells in
a quadrant have a similar field esteem. The method delivers a topsy turvy, tree like structure,
known as a quad tree. In primary memory, the hubs of a quad tree (the two circles and squares in
the figure beneath) are spoken to as records. The connections between them are pointers, a
programming method to address (for example to highlight) different records (Bak 1996,
Mandelbrot and Hudson 2004).
Quad trees are versatile on the grounds that they apply the spatial autocorrelation
guideline, for example that areas that are close in space are probably going to have comparable
field esteems. At the point when an aggregate of cells has a similar worth, they are spoken to
together in the quad tree, gave limits harmonize the predefined quadrant limits. This is the reason
we can likewise express that a quad tree gives a settled decoration: quadrants are possibly part in

Running Head: Application of GIS data and spatial thinking role in data science 7
the event that they have at least two qualities. The square hubs at a similar level speak to rise to
region sizes, permitting fast calculation of the zone related with some field esteem. The top hub
of the tree speaks to the total raster.
To outlines the above talk, we can say that decorations segment the investigation space
into cells and dole out an incentive to every cell (by a wide margin the most generally utilized).
The technique by which the investigation space is part into cells is somewhat subjective, as cell
limits normally have almost no bearing to this present reality marvels that are spoken to (Bak
1996, Mandelbrot and Hudson 2004).
Decorations don't expressly store geo-references of the wonders they speak to. Rather,
they give a geo-reference of the lower left corner of the raster, for example, in addition to a
marker of the raster€™s goals, in this manner certainly giving geo-references to all cells in the
raster. In vector portrayals, an endeavor is made to unequivocally relate geo-references with the
geographic wonders. A geo-reference is a facilitate pair from some geographic space, and is
otherwise called a vector. This clarifies the name. Underneath, we talk about different vector
portrayals. We start with our dialog with the TIN, a portrayal for geographic fields that can be
viewed as a half and half among decorations and vector portrayals (Batty, 2012).
A generally utilized information structure in GIS programming is the triangulated
unpredictable net-work, or TIN. It is one of the standard usage systems for computerized
territory models, however it tends to be utilized to speak to any consistent field. The standards
behind a TIN are straightforward. It is worked from a lot of areas for which we have an
estimation, for example a height. The areas can be discretionarily dissipated in space, and are

Running Head: Application of GIS data and spatial thinking role in data science 8
normally not on a pleasant customary lattice. Any area together with its height worth can be seen
as a point in three-dimensional space (Batty, 2012).
In three-dimensional space, three focuses exceptionally decide a plane, as long as they
are not collinear, for example they should not be situated on a similar line. A plane fitted through
these focuses has a fixed angle and inclination, and can be utilized to register a guess of height of
different areas. Since we can pick numerous triples of focuses, we can develop numerous such
planes, and in this way we can have numerous rise approximations for a solitary area.
In this way, it is shrewd to limit the utilization of a plane to the triangular territory
between the three focuses. On the off chance that we limit the utilization of a plane to the
territory between its three grapple focuses, we get a triangular decoration of the total
examination space. A few decorations are superior to other people, as in they make littler
mistakes of rise guess. For example, on the off chance that we base our height calculation for
area P on the left hand concealed triangle, we will get another incentive than from the correct
hand concealed triangle. The subsequent will give a superior estimate in light of the fact that the
normal good ways from P to the three triangle stays is littler (Bektas and Çöltekin 2012,
Cockburn et al. 2008, Çöltekin 2009).
The triangulation of the second figure above happens to be a Delaunay triangulation,
which it could be said is an ideal triangulation. There are various methods for characterizing
what such a triangulation is, yet we get the job done here to state two significant - properties. The
first is that the triangles are as symmetrical as they can be, given the arrangement of grapple
focuses. The subsequent property is that for every triangle, the circumcircle through its three stay

Running Head: Application of GIS data and spatial thinking role in data science 9
focuses doesn't contain some other grapple point. One such circumcircle is portrayed on the
privilege of the figure above).
A TIN unmistakably is a vector portrayal: each stay point has a put away geo-reference.
However, we may likewise consider it a sporadic decoration, as the picked triangulation gives a
parceling of the whole examination space. Be that as it may, for this situation, the cells don't
have a related put away an incentive as is ordinary of decorations, yet rather a basic insertion
work that uses the height estimations of its three stay focuses (Bektas and Çöltekin 2012,
Cockburn et al. 2008, Çöltekin 2009).
Focuses are characterized as single organize sets (x; y) when we work in 2D, or facilitate
triplets (x; y; z) when we work in 3D. The decision of arrange framework is another issue, which
we will talk about later. Focuses are utilized to speak to objects that are best depicted as shape-
and size-less, one-dimensional highlights. Regardless of whether this is the situation truly relies
upon the reasons for the spatial application and furthermore on the spatial degree of the articles
contrasted with the scale applied in the application. For a vacationer city map, a recreation center
won't normally be viewed as a point include, however maybe an exhibition hall will, and
unquestionably an open telephone stall may be spoken to as a point. Other than the geo-
reference, typically additional information is put away for each point object. This alleged
characteristic or topical information can catch whatever is viewed as applicable about the item.
For telephone stall questions, this may incorporate the owning phone organization, the telephone
number, or the information last adjusted. Line information are utilized to speak to one-
dimensional articles, for example, streets, railways, waterways, streams and electrical cables.
Once more, there is an issue of importance for the application and the scale that the application
requires. For the model utilization of mapping visitor data, transport, tram and streetcar courses

Running Head: Application of GIS data and spatial thinking role in data science 10
are probably going to be pertinent line highlights. Some cadastral frameworks, then again, may
believe streets to be two-dimensional highlights, for example having a width also (Bektas and
Çöltekin 2012, Cockburn et al. 2008, Çöltekin 2009).
Above, we talked about the thought that self-assertive, constant curvilinear highlights are
as similarly hard to speak to as consistent fields. GISs in this way estimated such highlights
(limitedly!) as arrangements of hubs. The two end hubs and at least zero inner hubs or vertices
characterize a line. Different expressions for direct that are generally utilized in some GISs are
plotline, curve or edge. A hub or vertex resembles a point (as examined above) yet it just serves
to characterize the line, and give shape so as to acquire a superior estimation of the real
component.
Discussion and Conclusion
It seems to me that all of the higher median household income is located in the center of
the Douglas County. I think this pattern is reasonable because most of the family located in the
center of Douglas county completed the college Education, therefore they more likely to make
more money compare to other family who doesn’t not attend to college, and in this case most of
the family who live far away from the center of Douglas County have a lower percent college
education compare to the family who live in the center of Douglas County. The percentage of
minority in the Douglas County tends to live to the East (right) of the center of Doulas County
because there is fewer minorities live in the left than the right of the Douglas County. I think the
reason why many minorities are living in the East of the map is because on the East of the map,
it towards the downtown direction, which have lots of African-American, and Mexican-
American people. A common way to think about the distribution of locations is to ask whether

Running Head: Application of GIS data and spatial thinking role in data science 11
they are dispersed (more evenly distributed than random), randomly distributed (locations cannot
be predicted because they could occur anywhere), or clustered (grouped together more than a
random distribution). The TRI facilities are clustered because most of the firms or corporation is
located on the East of the Douglas County; however there are a few firms that located on the
west of Douglas County. The relationship between the median household income and number of
TRI facilities are opposite, which mean that the fewer of the TRI facilities the higher of the
median household income, and vice versa (Bektas & Çöltekin 2012, Cockburn et al. 2008,
Çöltekin 2009).
I think the graph does not support my hypothesis, therefore rejected my hypothesis
because at some point on the graph, the higher of the median income (at $125,000) could also
have the more TRI facilities (60).

Running Head: Application of GIS data and spatial thinking role in data science 12
References
Agrawal, R., Imieliński, T., & Swami, A. (1993). Mining Association Rules between Sets of
Items in Large Databases. Proceedings of the 1993 ACM SIGMOD International
Conference on Management of Data, 22(2), 207-216. doi: 10.1145/170035.170072
Anthony, D., Henderson, T., & Kotz, D. (2007). Privacy in Location-Aware Computing
Environments. IEEE Pervasive Computing, 6(4), 64-72.
Bak, P. (1996). How Nature Works: The Science of Self-Organised Critically (1 ed.):
Copernicus.
Batty, M. (2012). Smart Cities, Big Data. Environmental and Planning B: Planning and Design,
39(2), 191-193. doi: 10.1068/b3902ed
Batty, M., & Longley, P. A. (1994). Fractal Cities: A Geometry of Form and Function. London:
Academic Press.
Bektas, K., & Çöltekin, A. (2012). Area of Interest Based Interaction and Geovisualization with
WebGL. Proceedings of The Graphical Web Conference 2012.
Bernasocchi, M., Çöltekin, A., & Gruber, S. (2012). An Open Source Geovisual Analytics
Toolbox for Multivariate Spatio-Temporal Data for Environmental Change Modeling.
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information
Sciences, 1-2(2), 123-128.
Bolbol, A., Cheng, T., & Haworth, J. (2011). Using a Moving Window SVMs Classification to
Infer Travel Mode from GPS Data. Proceedings of the 11th International Conference on
GeoComputation, 262- 270.

Running Head: Application of GIS data and spatial thinking role in data science 13
Carbin, M., Misailović, S., & Rinard, M. C. (2013). Verifying Quantitative Reliability for
Programs that Execute on Unreliable Hardware. ACM SIGPLAN NOTICES - OOPSLA
'13, 48(10), 33-52. doi: 10.1145/2544173.2509546
Castro-Neto, M., Jeong, Y.-S., Jeong, M.-K., & Han, L. D. (2009). Online-SVR for short-term
traffic flow prediction under typical and atypical traffic conditions. Expert Systems with
Applications 36, 6164– 6173. doi: 10.1016/j.eswa.2008.07.069
Chandola, V., Banerjee, A., & Kumar, V. (2009). Anomaly Detection: A Survey. ACM
Computing Surveys, 41(3), 1-58. doi: 10.1145/1541880.1541882
Chen, M., Mao, S., & Liu, Y. (2014). Big Data: A Survey. Mobile Networks and Applications,
19(2), 171-209. doi: 10.1007/s11036-013-0489-0
Cheng, T & Adepeju, M. (2014). Modifiable Temporal Unit Problem (MTUP) and its Effect on
Space-Time Cluster Detection. PLOS ONE 9(6), e100465.
Cheng, T., Haworth, J., Anbaroglu, B., Tanaksaranond, G., & Wang, J. (2014). Spatiotemporal
Data Mining. In M. M. Fischer & P. Nijkamp (Eds.), Handbook of Regional Science (pp.
1173-1193): Springer Berlin Heidelberg