Professional Documents
Culture Documents
by
Michaelangelo Severa
December 2023
Table of Contents
1. INTRODUCTION ................................................................................................ 1
1.1. Background and context of the study ............................................................ 1
1.2. Statement of the problem .............................................................................. 2
1.3. Purpose of the Study ..................................................................................... 3
1.4. Research Objectives ...................................................................................... 4
1.5. Significance of the study ............................................................................... 5
1.6. Scope of the Study ........................................................................................ 6
1.7. Limitations of the Study................................................................................ 8
1.8. Conclusion .................................................................................................... 9
2. LITERATURE REVIEW ................................................................................... 11
2.1. Crime Hotspots ........................................................................................... 12
2.1.1. Theories in Crime Mapping Research................................................. 14
2.1.2. Methods for Detecting Crime Hotspots .............................................. 15
2.1.3. Some Case Studies and Examples of Crime Hotspots in the Selected
Four Cities........................................................................................................... 17
2.1.4. Some Challenges and Future Directions in Crime Mapping............... 19
2.2. Major Factors Associated with Crime Patterns and Hotspots ..................... 22
2.2.1. Socioeconomic Factors ....................................................................... 24
2.2.2. Environmental Factors ........................................................................ 25
2.2.3. Spatial and Temporal Factors ............................................................. 27
2.2.4. Other Factors Associated with Crime Incidents.................................. 28
2.3. Understanding the Factors that Drive Crime Hotspots and Incidents in the
Philippines .............................................................................................................. 29
2.4. Synthesis and Identified Gaps in the Literature .......................................... 31
3. RESEARCH METHODOLOGY ........................................................................ 34
3.1. Data Collection ........................................................................................... 34
3.2. Data Preparation.......................................................................................... 35
3.2.1. Procedures for Enabling Spatial Attributes ......................................... 35
i
3.2.2. Discussion on Any Necessary Data Transformation or Standardization
36
3.3. Exploratory Data Analysis .......................................................................... 37
3.4. Spatial Autocorrelation Analysis – Global Moran’s I................................. 38
3.5. Hotspot Analysis ......................................................................................... 40
3.6. Cluster and Outlier Analysis ....................................................................... 42
3.7. Geographically Weighted Regression Analysis .......................................... 44
3.8. Space-Time Mining .................................................................................... 46
3.9. Data Interpretation and Reporting............................................................... 48
3.10. Ethical Considerations ............................................................................ 50
3.11. Conclusion .............................................................................................. 52
4. Bibliography ....................................................................................................... 54
ii
1. INTRODUCTION
Quezon City, the most populous city in the Philippines, has experienced
fluctuations in crime rates from 2019 to early 2023 (Quezon City Local Government,
2023). These fluctuations highlight the importance of studying the spatial distribution
of crime and its socioeconomic correlates to better understand the factors contributing
to crime in the city. The e-Crime Incident Recording and Analysis System (e-CIRAS)
records and analyzes crime incidents in Quezon City, providing valuable data for
its socioeconomic correlates, researchers and policymakers can better understand the
activity (Thomson, Espin, & Samuels-Jones, 2020). Some studies have used
Geographic Information Systems (GIS) to map and analyze crime patterns in the
GIS to analyze property crimes such as carnapping, robbery, and theft incidents
been found to be correlated with crime rates. In general, lower socioeconomic status
is associated with higher crime rates. By examining the spatial distribution of crime
1
and its socioeconomic correlates in Quezon City, researchers can identify patterns and
Some studies have used spatial analysis techniques to identify crime hotspots
and clusters. For example, the nearest neighbor hierarchical clustering technique has
been used to identify groups of incidents that are spatially close. Spatial regression
analysis has also been used to explore the relationship between crime rates and
understand the factors contributing to crime in the area. This could involve using GIS
to map crime incidents, exploring the relationship between crime rates and
socioeconomic factors, and identifying crime hotspots and clusters. By doing so,
address the underlying causes of crime and improve public safety in Quezon City.
strategies.
1. What are the spatial patterns of crime in Quezon City, and how do they
relate to the socioeconomic factors in the area?
2
2. Are there identifiable crime hotspots and clusters in Quezon City, and if
so, what are their characteristics?
contributing to crime in the city by examining the spatial distribution of crime and its
relationship with socioeconomic factors. This understanding will help inform the
3
The study will contribute to the existing body of knowledge on spatial
clustering of crime and its socioeconomic correlates. The findings will also be
valuable for policymakers and law enforcement agencies in Quezon City, as they can
use this information to develop targeted interventions and effective crime prevention
strategies to address the underlying causes of crime and improve public safety in the
city.
The research objectives for this study on Spatial Clustering of Crime and Its
1. To identify and characterize crime hotspots in the city and determine the
spatial clustering of specific crime types.
2. To investigate the association between high unemployment rates in
neighborhoods and the prevalence of crime hotspots.
3. To analyze temporal changes in the geographic distribution of crime and
detect the emergence of new hotspots.
4. To examine the presence of local spatial outliers in crime rates, such as
contrasting crime rates in neighboring areas.
5. To explore the influence of factors like income levels, education levels,
and proximity to public transportation on the spatial distribution of crime.
These research objectives aim to provide a comprehensive understanding of
the complex interplay between crime and its socioeconomic correlates in Quezon
City. By uncovering patterns and relationships between these factors, this research
will contribute to enhancing public safety and fostering a more secure environment
4
1.5. Significance of the study
The significance of this research study lies in its dual contribution to both the
focusing on the examination of the spatial distribution of crime and its intricate
relationship with various socioeconomic factors, the study is poised to enrich the
current body of knowledge pertaining to the complex dynamics between crime and its
socioeconomic determinants.
From a theoretical perspective, the study aims to delve deeper into the
contributing factors that lead to spatial clustering of crime and its ties with critical
employment status. The study will extrapolate data and draw connections that can
describe the mechanics of crime in urban environments. This exploration has the
potential to pave the way for novel perspectives and hypotheses about urban crime
subject matter.
serve as a crucial resource for policymakers and law enforcement agencies in Quezon
socioeconomic influences that exacerbate these hotspots, the study offers a unique
opportunity for these stakeholders to devise targeted interventions and potent crime
5
prevention strategies. Such strategies, informed by a clear understanding of the
Additionally, the insights from this study can shape the design and execution
of evidence-based policies and initiatives that directly address the root causes of
crime, such as social and economic disparities. By highlighting the need for strategies
that not only tackle crime but also promote community cohesion and resilience, the
boundaries of Quezon City. Law enforcement and public safety in the city are
maintained and controlled by the Philippine National Police (PNP), specifically the
Quezon City Police District (QCPD). The QCPD oversees police functions in the
area, which is divided into 16 Police Stations with component Police Community
Precincts. Their main objective is to enforce the law and prevent and control crimes,
Quezon City between January 2020 and March 2023. This period represents a recent
timeframe that could provide relevant and updated data on crime trends in the area. It
is essential to analyze crime patterns within this specific timeframe to ensure the
6
The study will focus on the following research questions and their
1. Are there areas of the city where high crime rates are not random but
form specific clusters (hotspots)?
- Crime incident counts per geographic unit (e.g., census tract, postal
code), normalized by the population size in each area to get crime
rates per unit of population.
4. How does the geographic distribution of crime change over time? Are
there emerging hotspots?
- Crime incident counts per geographic unit for multiple time periods
(e.g., monthly, yearly), normalized by the population size in each
area, if possible.
5. Are there local spatial outliers in crime rates, i.e., areas of high crime
rates surrounded by areas of low crime rates (or vice versa)?
- Crime incident counts per geographic unit, normalized by the
population size in each area.
7
1.7. Limitations of the Study
The proposed study has several limitations that may affect the reliability and
First, the study focuses on a specific geographical area, Quezon City, which
may limit the generalizability of the findings to other cities or regions. However, the
study's focus on Quezon City allows for a detailed examination of crime patterns and
Second, the study considers crimes committed in the City between January
2020 and March 2023, which may not capture long-term trends or historical patterns
and updated data on crime trends in the area, ensuring the relevance of the study's
Third, the study relies on measurable indicators and available data for crime
and socioeconomic factors, which may not capture all relevant factors influencing
crime patterns in Quezon City. For example, unmeasured factors such as social
cohesion, informal social control, and neighborhood characteristics may also play a
Lastly, the dataset used in the study was collected during the pandemic,
which may have influenced the crime scenario in Quezon City due to lockdown
measures and other restrictions. As cities are now reopening for business, the crime
8
patterns and trends may have changed, and the findings from the study may not fully
limited range of crime types, relies on measurable indicators, and uses data collected
during the pandemic, it still provides important insights into the spatial patterns of
patterns and relationships, the study can contribute to the development of more
communities in Quezon City and potentially informing similar research in other urban
areas.
1.8. Conclusion
The background and context of the study on Spatial Clustering of Crime and
Its Socioeconomic Correlates in Quezon City, Philippines, have been provided. The
statement of the problem has been identified, emphasizing the need for a
with socioeconomic factors. The purpose of the study and the research objectives
have been outlined, focusing on the investigation of crime hotspots, the association
between high unemployment rates and crime, temporal changes in crime distribution,
local spatial outliers, and the influence of factors like income levels, education levels,
and proximity to public transportation on crime patterns. The research questions that
will guide the study, the significance of the study in terms of its contribution to both
9
theory and practice, the scope of the study in terms of geographical and time
boundaries, and the limitations of the study, including the focus on a specific
geographical area, the selected timeframe, reliance on measurable indicators, and the
Having established the foundation of the study in this introduction, the next
chapter will delve into a review of the relevant literature on spatial clustering of crime
and its socioeconomic correlates. This literature review will provide a comprehensive
understanding of the existing body of knowledge on the subject, identify gaps in the
previous research and theories on crime patterns, socioeconomic factors, and spatial
analysis techniques, the literature review will further contextualize the study and
provide a solid basis for the subsequent chapters on methodology, data analysis, and
10
2. LITERATURE REVIEW
existing body of knowledge on the subject. The review will be divided into two main
sections: the concept of crime hotspots and major factors associated with crime
In the first section, the concept of crime hotspots will be explored, including
methods for detecting crime hotspots, such as hotspot mapping, clustering, spatial
scan statistics, Geographic Information System (GIS) software, Spatial and Temporal
Additionally, case studies and examples of crime hotspots in selected cities, including
New York City, Los Angeles, Chicago, and London, will be presented. The section
will conclude with a discussion of challenges and future directions in crime mapping,
including varying factors in hotspot analysis, the gap between theory and practice,
In the second section, major factors associated with crime patterns and
hotspots will be examined. This section will cover socioeconomic factors such as
factors like land use and zoning, proximity to public transportation, and presence of
11
public spaces. Spatial and temporal factors, including seasonal and time-of-day
patterns and geographic distribution of crime events, will also be discussed. Other
factors associated with crime incidents, such as demographic characteristics (e.g., age,
Throughout this section, previous studies and methodologies used to identify these
often depicted on maps. These areas are useful for researchers and analysts to study
the geography of crime (Crime hotspots, 2023). There are two ways to interpret crime
hotspots. To start, certain areas with higher crime rates are often referred to as
clustering can be employed to pinpoint these areas. (He, Lai, Wang, Liu, & Deng,
make generalizations about both hotspot areas and the areas surrounding them (Crime
hotspots, 2023).
Studies have consistently shown that crime in cities tends to occur in specific
geographic areas, with most streets being relatively safe and only a small percentage
of blocks experiencing high levels of criminal activity. These areas are commonly
referred to as "hot spots," and can be identified using Geographic Information System
12
(GIS) software to track crime rates. Although hot spots can be a persistent issue, they
Being able to identify crime hotspots is essential for enhancing public safety
measures. Through hot spot analysis, law enforcement can pinpoint areas with high
rates of crime, the types of criminal activity occurring, and the most effective
strategies for response (Gonzales, Schofield, & Hart, 2005). To conduct a thorough
hot spot analysis, there are several key factors to consider. These include the nature of
the analysis, spatial relationships, the type of crime being studied, time intervals,
potential barriers, the method of output display, and the software utilized. (Gonzales,
Schofield, & Hart, 2005). To avoid using arbitrary methods in hot spot analysis,
analysts should prioritize establishing a solid connection between theory and practice,
relying on a scientific foundation. This approach ensures more effective and accurate
This section will explore the concept of crime hotspots by discussing some
theoretical ideas of crime mapping, then the methods to which these theoretical
perspectives are applied. This paper will also present case studies to demonstrate the
and developing targeted policing strategies in the selected four major cities. Its
concluding part will also mention the challenges and future directions in crime
mapping, including the need to address varying factors in hot spot analysis and
13
2.1.1. Theories in Crime Mapping Research
spatial patterns in criminal activity by utilizing location data about crime events
(Crime Mapping, 2023). As crime mapping tools and techniques continue to evolve,
crime (Crime Mapping, 2023). This section will provide an overview of several
14
associated with higher levels of criminal involvement, crime mappers can
develop effective strategies for crime prevention.
To understand and prevent crime, crime mapping is a crucial tool. By
analyzing the spatial and temporal patterns of criminal activity, as well as the social,
psychological, and economic factors that contribute to it, crime mappers can identify
the root causes of crime hotspots and implement effective prevention strategies. In the
following section, we will explore various techniques for identifying areas with high
crime rates.
activity, known as crime hotspots. Detecting these hotspots involves using various
methods that depend on the type of analysis required. This section outlines several
techniques that can be used to identify crime hotspots and develop effective crime
prevention strategies.
15
intensity, even if they are not directly adjacent to each other (He, Lai,
Wang, Liu, & Deng, 2022).
Several methods such as hotspot mapping, clustering, spatial scan statistics, GIS
16
software, STAC, and nearest neighbor prove to be useful in detecting crime hotspots.
By utilizing these techniques, crime mappers can pinpoint areas with high crime rates
and devise effective strategies for preventing them. The next section will delve into
Maps can be used to visualize crime hotspots, which refer to areas with a
researchers and analysts can examine geographic areas of crime and develop effective
strategies for crime prevention. Several cities have utilized crime mapping to identify
these hotspots and develop successful crime prevention strategies. In this section, we
four cities.
17
Overall, the study utilized spatio-temporal analysis to predict criminal
patterns in NYC neighborhoods. The XGBoost classifier outperformed
other models, making it a suitable option for detecting crimes in
neighborhoods. (Almuhanna, Alrehili, Alsubhi, & Syed).
2. Los Angeles
The Almanie et al. 2015 study delves into identifying spatial and
temporal criminal hotspots, with a particular focus on Denver and Los
Angeles. By analyzing crime data and utilizing data-mining models, the
report aims to pinpoint criminal hotspots in these areas. The Los Angeles
Crimes Dataset contains 14 attributes and 243750 instances, with 96% of
the crimes occurring in 2014. The report successfully achieved its goal by
utilizing the Apriori algorithm on both datasets to identify criminal
hotspots. In Los Angeles, the areas with the highest likelihood of crime
are 77th Street, Southwest, Pacific, N Hollywood, Southeast, Northeast,
and Van Nuys. The most frequent occurrence of crime takes place in 77th
Street on Mondays between 9 pm and 1 am. Crimes in both 77th Street
and Southwest areas occur every day, mostly between 8 am and midnight.
These findings can be useful in raising awareness about dangerous
locations and aiding agencies in predicting future crimes in specific
locations at particular times. (Almanie, Mirza, & Lor).
3. Chicago
4. London
18
clusters of crime based on different denominators (Malleson & Andresen,
Exploring the Impact of Ambient Population Measures on London Crime
Hotspots, 2016).
various cities for identifying crime hotspots and creating effective crime prevention
strategies. Through hotspot mapping, clustering, spatial scan statistics, GIS software,
STAC, and nearest neighbor, crime mappers can detect high crime areas and devise
successful solutions. It's worth mentioning that crime mapping has been implemented
in major cities like New York City, Los Angeles, Chicago, and London, resulting in
Mapping
Crime mapping has become a vital tool in preventing crime and has been
adopted by numerous cities to identify high-crime areas and develop effective crime
prevention strategies. Although it is valuable, there are still areas that require
addressed include:
19
1. Varying Factors in Hot Spot Analysis
20
Despite these challenges, the adoption of GIS-based crime analysis tools by
the Philippine police, such as the La Trinidad Municipal Police Station, has shown
World, 2012). These tools can help map crime density and inform appropriate actions
technologies and invest in training local police personnel in crime mapping and
planning, the potential for effective crime prevention and management through
include the use of machine learning algorithms and social media data to identify crime
hotspots and patterns (Shah, Nandish, & Manan, 2021) & (Wang, Yu, Liu, & Young,
2019). Machine learning algorithms can analyze large amounts of data and identify
patterns that may not be easily detected by humans (Shah, Nandish, & Manan, 2021).
Social media data can provide real-time information about crime events and help
crime mappers identify areas with high crime intensity (Shah, Nandish, & Manan,
2021), & (Bendler, Ratku, & Neumann, 2014). These advancements offer new
opportunities for improving the accuracy and effectiveness of crime mapping and
prevention strategies.
Future directions in crime mapping and analysis may include the integration
and identify crime hotspots. Additionally, the use of social media data, such as
21
geolocated Twitter messages, can help researchers better understand the spatial
distribution of crime and the population at risk (Malleson & Andresen, Using Social
Media Data to Assess Spatial Crime Hotspots), (Wang, Yu, Liu, & Young, 2019).
These emerging trends in crime mapping and analysis can contribute to a more
and factors, ultimately informing the development of more effective crime prevention
strategies and interventions (Shah, Nandish, & Manan, 2021), (Wang, Yu, Liu, &
Young, 2019), (Saeed & Abdulmohsin, 2023), (Yadav, Kumar, Bhatnagar, & Verma,
2019).
To sum up, crime mapping has become a crucial means of preventing crime.
It has been utilized in different cities to pinpoint crime hotspots and devise efficient
strategies for preventing crime. Nonetheless, there are still challenges and future
directions that need to be tackled in crime mapping. By addressing these issues and
exploring future possibilities, crime mappers can devise even more effective strategies
Hotspots
Crime patterns and hotspots refer to the spatial and temporal distribution of
effective crime prevention strategies and allocate resources efficiently. High crime
concentrations characterize these hotspots and are often visualized using maps to help
22
researchers and analysts examine geographic areas of crime (Crime hotspots, 2023).
There are different techniques that have been devised to identify areas with high
crime rates, such as clustering and statistics-based methods (He, Lai, Wang, Liu, &
Deng, 2022). Identifying the factors associated with crime hotspots is essential for
environmental, spatial, and temporal aspects, researchers can gain insights into the
complex dynamics that contribute to the formation of crime hotspots. This knowledge
can then inform evidence-based policies and practices to reduce crime and improve
The subsequent sections of this study will delve deeper into the significant
environmental, spatial, and temporal factors and other factors. These sections will
also discuss the methodologies used in previous studies to identify these factors,
between the distinct aspects and how they may interact to influence crime patterns.
This integrated approach will enable us to understand the complex nature of crime
hotspots better and inform more effective crime prevention strategies. Understanding
the factors associated with crime hotspots is crucial for developing targeted
and their interactions, we can gain valuable insights into the complex dynamics that
23
contribute to the formation of crime hotspots and inform evidence-based policies and
substantial. In this part of the paper, we'll delve into the effects of different
unemployment rates. Furthermore, we'll examine earlier research and techniques used
1. Income Levels
2. Education Levels
3. Unemployment Rates
24
criminal behavior. Additionally, unemployed individuals may have more
time, increasing the likelihood of engaging in criminal activities.
significant and cannot be disregarded. This outline will explore the influence of
various environmental factors on crime, such as land use and zoning, proximity to
public transportation, and public spaces like parks and schools. Additionally, we will
review past studies and methodologies used to identify and analyze these crime
factors.
The way land is used and zoned has a big impact on where crime tends to
occur. Certain types of land use, such as alcohol outlets, clubs and discos,
cultural facilities, municipal housing, and commercial buildings, are
known to attract more crime than others (Sypion-Dutkowska & Leitner,
25
2017). These land use types can create opportunities for criminal
activities and contribute to the formation of crime hotspots. Conversely,
other land use types, such as residential areas with solid social
organization and stability, may be associated with lower crime rates
(Sackett, 2023).
The design of public spaces, like streets and parks, can affect crime rates
by influencing the interactions between offenders, potential victims, and
supervisors. The environmental features within these spaces play a crucial
role in determining the timing and location of criminal activity (Huafang,
Liu, & Yue, 2022). According to the routine activity theory, street crimes
are influenced by environmental factors, and these crimes typically take
place on the streets (Huafang, Liu, & Yue, 2022). To reduce the risk of
crime, it's crucial to comprehend how the street environment and spatial
changes affect criminal activity (Huafang, Liu, & Yue, 2022).
Additionally, public spaces can provide opportunities for offenders to
observe potential victims and plan their crimes, making these areas more
susceptible to criminal activities.
26
mixed land use and urban stability (Gu, Newman, Park, Lee, & Kim,
2019).
shape crime patterns and hotspots. This section of the outline will examine the impact
of different spatial and temporal factors on crime, such as seasonal and time-of-day
patterns, the distribution of crime events across geographic locations, and previous
studies that have identified these factors. For instance, a study conducted in the
more likely to happen in the afternoon. These incidents are frequently observed in the
road networks of the central business district (Taclay & Taclay, 2022).
Seasonal patterns in crime rates have been widely studied, with many
types of crimes showing fluctuations throughout the year (Lauritsen &
White, 2014). Environmental factors, such as temperature changes and
daylight hours, can influence crime rates, as can common activity patterns
associated with the beginning and end of the school year (Lauritsen &
White, 2014). As an example, incidents of household burglary and
larceny are often more prevalent during the summer months and decrease
during other seasons. However, motor vehicle thefts do not follow the
same seasonal trends (Lauritsen & White, 2014). Violent victimizations
also show seasonal patterns (Lauritsen & White, 2014).
27
weekends and that this temporal change has a spatial component
(Andresen & Malleson, 2015).
in shaping crime patterns and hotspots. This section directs the impact of these factors
on crime.
28
Research has shown that there is a correlation between certain
demographic factors and crime rates. For instance, younger individuals
have a higher likelihood of engaging in criminal activities. Furthermore,
males and certain racial groups are perceived to account for a larger
portion of offenders (Gramlich, 2020). Low-income people are much
more likely than others to experience violent crime (Sackett, 2023). These
disparities may be influenced by neighborhood characteristics and
socioeconomic factors (Sackett, 2023).
Various studies have explored the correlation between crime rates and
demographic factors like age, gender, and race in the context of social
disorganization. According to one study, crime rates tend to be higher
among certain racial groups, younger individuals, and males (Gramlich,
2020). Another study on serious and violent juvenile offenders
highlighted the differences in offending rates based on age, gender, and
race (Loeber & Farrington, 1998). In terms of social disorganization and
community cohesion, research has shown that age, gender, and race are
characteristics that can influence crime rates in certain areas (Schaefer,
2008). A study on criminal homicides in a prominent northern city
collected data on demographic factors and found associations between
these factors and crime rates.
The interplay of factors associated with crime patterns and hotspots in the
temporal, and spatial factors. Understanding these factors is crucial for developing
29
effective crime prevention strategies and law enforcement interventions in the
country.
The Philippines' crime patterns and hotspots are largely shaped by socio-
economic factors like poverty, unemployment, and education. These factors affect the
spatial distribution of crime, with areas experiencing high levels of poverty and
unemployment being more prone to criminal activity. Crime opportunity factors, such
contribute to the formation of crime hotspots. It's crucial to consider these factors
Temporal factors, such as the variation in crime rates throughout the day,
week, and year, also influence crime patterns and hotspots. Understanding these
temporal patterns can help law enforcement agencies allocate resources more
effectively and efficiently. Spatial factors, including land use, urban design, and
these factors in the context of the Philippines can provide valuable insights into the
to various socio-economic and crime opportunity factors (Taclay & Taclay, 2022).
Another study detected potential illegal drug trafficking hotspots in Manila using a
30
fuzzy clustering approach (Aycardo & Tubay, 2017). These studies highlight the
and policymakers can develop more effective crime prevention strategies and
The literature review has explored two main themes: the concept of crime
hotspots and major factors associated with crime patterns and hotspots. This synthesis
will discuss how these studies collectively contribute to the understanding of these
provided valuable insights into the theories in crime mapping research, methods for
detecting crime hotspots, and case studies and examples of crime hotspots in selected
patterns of crime and the factors that drive the formation of crime hotspots. However,
gaps remain in the literature, particularly in terms of the varying factors in hotspot
analysis, the gap between theory and practice, perceived challenges in conducting
hotspot mapping in the Philippines, and future directions in crime mapping. Further
research is needed to address these gaps and develop more effective methods for
31
Major Factors Associated with Crime Patterns and Hotspots. The
literature on major factors associated with crime patterns and hotspots has examined
the role of socioeconomic factors, environmental factors, spatial and temporal factors,
shaping crime patterns. These studies have provided valuable insights into the
complex interplay between these factors and crime, informing the development of
remain in the literature, particularly in terms of understanding the factors that drive
crime hotspots and incidents in the Philippines. Further research is needed to explore
the unique context of Quezon City and identify the specific factors that contribute to
the spatial clustering of crime and its socioeconomic correlates in this urban setting.
and its socioeconomic correlates. However, gaps and unresolved issues remain in the
and the factors that drive crime hotspots and incidents in the Philippines.
With these unresolved issues, the study aims to address these gaps in the
following ways:
1. Varying factors in hotspot analysis: The study will examine the specific
factors contributing to the formation of crime hotspots in Quezon City,
considering the unique context of the city and exploring the interaction
between various factors, such as socioeconomic and environmental
variables.
32
2. The gap between theory and practice: By focusing on the spatial
clustering of crime and its socioeconomic correlates in Quezon City, the
study aims to provide practical insights that can inform the development
of targeted interventions and effective crime prevention strategies for
policymakers and law enforcement agencies in the city.
33
3. RESEARCH METHODOLOGY
The primary data source for this study will be the Crime Information,
Reporting, and Analysis System (CIRAS) provided by the Philippine National Police
types of crime, and their corresponding geographic locations within Quezon City.
The criteria for data selection in this study include the following:
2. Timeliness: The data should cover the specified timeframe of the study,
which is from January 2020 to March 2023.
Data collection procedures for this study will involve the following steps:
1. Access the CIRAS data source provided by the Philippine National Police
(PNP).
3. Organize the extracted data into a suitable format for analysis, such as a
spreadsheet or a Geographic Information System (GIS) software.
34
5. Collect additional data on socioeconomic factors (e.g., income levels,
education levels, unemployment rates), environmental factors (e.g., land
use and zoning, proximity to public transportation, presence of public
spaces), and other factors (e.g., demographic characteristics, social
disorganization, community cohesion) from relevant sources, as needed,
to address the research questions and objectives.
The data preparation process for the study involves several steps to ensure
that the data is clean, well-formatted, and suitable for analysis. After accessing the
CIRAS data source provided by the Philippine National Police (PNP) and extracting
relevant data on crime incidents, types of crime, and their corresponding geographic
locations within Quezon City for the specified timeframe (January 2020 to March
followed:
35
geocoding techniques, which involve matching the location information
(e.g., street addresses, postal codes) to a spatial reference system.
2. Spatial data integration: The geocoded crime data will be integrated with
other spatial data sources, such as census tracts or postal codes, to enable
the analysis of crime patterns in relation to various geographic units and
socioeconomic factors.
3. Spatial data projection: The spatial data will be projected into a suitable
coordinate system (e.g., Universal Transverse Mercator, Philippine
Reference System) to ensure accurate representation and measurement of
spatial relationships in the study area.
Standardization
data is suitable for analysis and to facilitate the comparison of variables with different
36
3.3. Exploratory Data Analysis
statistical analysis and data visualization techniques will be employed to gain a better
understanding of the data and identify any patterns or trends. These techniques may
include:
The preliminary findings from the exploratory data analysis will provide
insights into the overall patterns and trends in the data, as well as any potential
37
2. Observations of relationships between crime rates and socioeconomic
factors, such as income levels, education levels, and unemployment rates,
indicate potential associations between these factors and crime patterns.
variable of interest (e.g., crime rates) is correlated with itself across space. A positive
Global Moran's I value indicates that similar values tend to be located near each other,
that dissimilar values are located near each other, suggesting a dispersed or
To compute the Global Moran's I statistic, the following steps are typically followed:
2. Calculate the deviation of each observation (e.g., crime rate) from the
mean value of the variable.
3. Multiply the deviations by the corresponding spatial weights and sum the
products.
38
4. Divide the sum of the products by the sum of the squared deviations from
the mean.
The interpretation of the Global Moran's I results depends on the value of the
statistic and its significance level (e.g., p-value). A statistically significant positive
value indicates the presence of spatial clustering or hotspots in the variable of interest
(e.g., crime rates), suggesting that areas with high (or low) values are more likely to
be located near each other. This finding can inform the identification of crime
strategies.
checkerboard-like pattern, suggesting that areas with high values are more likely to be
located near areas with low values, and vice versa. This finding can inform the
space. This finding suggests that the spatial distribution of the variable may not be
39
can help assess the overall spatial correlation between crime rates and other factors,
providing insights into the spatial processes and factors that may be driving the
degree of spatial clustering of high or low values for a given variable (e.g., crime
statistically significant positive Gi* value indicates a local hotspot, where high values
are clustered together. A statistically significant negative Gi* value indicates a local
coldspot, where low values are clustered together. A non-significant Gi* value
To conduct the hotspot analysis using the Getis-Ord Gi* statistic, the
2. Calculate the Gi* statistic for each geographic unit, taking into account
the values of the variable of interest (e.g., crime rates) and the spatial
weights matrix.
40
The hotspot analysis can be conducted using various software tools and
The interpretation of the Getis-Ord Gi* results depends on the values of the
positive Gi* values indicate the presence of local hotspots, where high values of the
variable of interest (e.g., crime rates) are clustered together. These hotspots can
inform the identification of areas with high crime rates and the development of
coldspots, where low values of the variable are clustered together. These coldspots
can inform the identification of areas with low crime rates and the development of
Non-significant Gi* values suggest that the values in the area are randomly
distributed, indicating that the spatial distribution of the variable may not be
In consideration of this study, the hotspot analysis using the Getis-Ord Gi*
statistic can help identify areas with high crime rates or specific types of crime
clustered locations, providing insights into the spatial processes and factors that may
41
3.6. Cluster and Outlier Analysis
(LISA) used to identify local spatial clusters and outliers in a dataset. It measures the
degree of spatial autocorrelation for a given variable (e.g., crime rates) in a specific
significant positive Local Moran's I value indicates a local cluster, where high values
are surrounded by high values (HH) or low values are surrounded by low values (LL).
outlier, where a high value is surrounded by low values (HL) or a low value is
that the values in the area are randomly distributed, indicating no significant local
spatial autocorrelation.
To conduct the cluster and outlier analysis using the Anselin Local Moran's I
statistic, the following steps are typically followed:
1. Define a spatial weights matrix, which represents the spatial relationships
between the geographic units (e.g., census tracts, postal codes) in the
study area. This matrix can be based on various criteria, such as
contiguity, distance, or inverse distance.
2. Calculate the Local Moran's I statistic for each geographic unit, taking
into account the values of the variable of interest (e.g., crime rates) and
the spatial weights matrix.
and packages, such as ArcGIS, GeoDa, or R (e.g., using the "spdep" package) .
42
The interpretation of the Anselin Local Moran's I results depends on the
values of the statistic and their statistical significance (e.g., p-values). Statistically
significant positive Local Moran's I values indicate the presence of local clusters (HH
or LL), suggesting that areas with similar values of the variable of interest (e.g., crime
rates) are located near each other. These clusters can inform the identification of areas
with high or low crime rates and the development of targeted interventions and crime
prevention strategies.
of local spatial outliers (HL or LH), revealing areas that are unusually safe or
the identification of areas with unique crime patterns and the development of
Non-significant Local Moran's I values suggest that the values in the area are
randomly distributed, indicating that the spatial distribution of the variable may not be
The cluster and outlier analysis using the Anselin Local Moran's I statistic can
help identify local spatial clusters and outliers in crime rates, providing insights into
the spatial processes and factors that may be driving the observed crime patterns in
the city.
43
3.7. Geographically Weighted Regression Analysis
that allows for the modeling of spatially varying relationships between a dependent
variable (e.g., crime rates) and one or more independent variables (e.g., income levels,
models, which assume that the relationships between variables are constant across
space, GWR allows for the estimation of local regression coefficients that can vary
across the study area. This enables the identification of spatially varying relationships
and provides a more nuanced understanding of the factors influencing the dependent
heterogeneity in the relationships between variables, which can lead to more accurate
relationships, GWR can inform the development of targeted interventions and policies
that address the specific factors driving the dependent variable in different areas.
2. Specify the GWR model: Define the dependent variable (e.g., crime
rates) and the independent variables (e.g., income levels, education levels,
proximity to public transportation) to be included in the model. Select an
appropriate spatial weights matrix to represent the spatial relationships
between the geographic units in the study area.
44
3. Estimate the GWR model: Use a GWR software or package to estimate
the local regression coefficients for each geographic unit, taking into
account the spatial weights matrix and the specified model.
4. Assess the model fit and diagnostics: Evaluate the goodness-of-fit of the
GWR model using measures such as the adjusted R-squared, the Akaike
Information Criterion (AIC), or the Bayesian Information Criterion
(BIC). Examine the diagnostics of the model, such as the residuals and
the local regression coefficients, to identify potential issues or areas for
improvement.
GWR can be conducted using various software tools and packages, such as
spatially varying relationships between the dependent variable (e.g., crime rates) and
the independent variables (e.g., income levels, education levels, proximity to public
negative relationship with the dependent variable in that specific location. Non-
significant local regression coefficients suggest that the independent variable does not
possible to identify areas where specific factors have a stronger or weaker influence
on the dependent variable. This information can inform the development of targeted
interventions and policies that address the specific factors driving the dependent
45
The GWR analysis can help identify spatially varying relationships between
crime rates and factors such as income levels, education levels, and proximity to
understanding of the spatial processes and factors driving crime patterns in the city
and inform the development of more effective crime prevention strategies and
interventions.
patterns in data that evolve over both space and time. These techniques are
particularly useful for studying phenomena, such as crime rates, that exhibit spatial
over multiple time periods, space-time pattern mining can help identify emerging
hotspots, trends, and changes in crime patterns, providing valuable insights for
(e.g., latitude and longitude) and one temporal dimension (e.g., time). Within the
followed:
46
1. Prepare the data: Organize the crime incident counts per geographic unit
for multiple time periods (e.g., monthly, yearly), normalized by the
population size in each area, if possible. Ensure that the data is clean,
well-formatted, and spatially and temporally referenced.
Space-time pattern mining can be conducted using various software tools and
packages).
examining the identified patterns and trends within the space-time cube to gain
insights into the spatial and temporal dynamics of crime in the study area. Some
possible findings from the space-time pattern mining analysis may include:
time in specific locations, suggesting the need for targeted interventions and crime
prevention strategies. Detection of changes in crime patterns over time, such as shifts
in the spatial distribution of crime incidents or the types of crime that are more
prevalent in certain areas. Insights into the relationships between crime rates and other
transportation, over space and time, informing the development of more effective and
47
In short, the space-time pattern mining analysis can help identify how the
geographic distribution of crime changes over time and reveal any emerging hotspots
understanding of the spatial and temporal processes driving crime in the city and
interventions.
The interpretation of the results of the spatial analyses conducted in the study
between crime rates and various socioeconomic factors. This process includes
assessing the statistical significance of the findings, as well as considering the spatial
48
4. Consider the temporal context of the data, such as changes in crime
patterns over time or the emergence of new hotspots, to identify trends
and inform the development of targeted interventions and crime
prevention strategies.
findings from the various spatial analyses and considering their implications for the
The findings of the study were reported in a clear, concise, and well-
communicate the results and their implications. The reporting process typically
involves:
49
2. Describing the patterns, trends, and relationships observed in the data,
and explaining their relevance to the research questions and objectives.
By following these steps, the study ensures that the findings are
2. Data security: The storage, processing, and sharing of crime data should
be conducted in a secure manner to prevent unauthorized access,
disclosure, or misuse of the information. This may involve using
password-protected files, encrypted storage systems, or secure data
transfer protocols.
50
3. Data accuracy and quality: The accuracy and quality of the crime data
should be carefully assessed and maintained to ensure that the findings
and conclusions of the study are based on reliable information. This may
involve verifying the data sources, addressing any inconsistencies or
errors in the data, and documenting any limitations or uncertainties
associated with the data.
4. Ethical use of data and findings: The findings and conclusions of the
study should be used responsibly and ethically, with the aim of informing
the development of more effective crime prevention strategies and
interventions. The study should avoid sensationalizing or stigmatizing
specific areas, communities, or individuals based on the crime patterns
identified, and should instead focus on understanding the underlying
spatial processes and factors driving these patterns.
following ways:
51
4. Integrity and transparency. Crime researchers should conduct their
studies with honesty, integrity, and transparency, ensuring that the
research methods, data, and findings are accurately and objectively
reported. This includes disclosing any potential conflicts of interest,
limitations, or uncertainties associated with the research, as well as being
open to scrutiny and feedback from peers, stakeholders, or the public. For
example, researchers should clearly describe the data sources, analytical
methods, and assumptions used in their study, and provide a balanced
interpretation of the results, acknowledging any limitations or alternative
explanations.
3.11. Conclusion
In this chapter, various steps and techniques involved in conducting the study
Philippines were discussed. The chapter covered data collection, data preparation,
research.
The next chapter, "IV. Results and Discussion," will present the findings of
the spatial analyses conducted in the Methodology chapter. It will discuss the
52
identified patterns, trends, and relationships between crime rates and various
socioeconomic factors in Quezon City, Philippines. The chapter will also provide a
detailed interpretation of the results, drawing on the techniques and insights from the
Methodology chapter to explain the observed patterns and their implications for crime
prevention strategies and interventions. Furthermore, the next chapter will discuss the
limitations of the study and suggest directions for future research in the field of spatial
crime analysis.
53
4. Bibliography
Almanie, T., Mirza, R., & Lor, E. (n.d.). Crime Prediction Based On Crime Types
And Using Spatial And Temporal Criminal Hotspots. International Journal of
Data Mining & Knowledge Management Process, 5(4).
doi:https://doi.org/10.48550/arXiv.1508.02050
Almuhanna, A. A., Alrehili, M. M., Alsubhi, S. H., & Syed, L. (n.d.). Prediction of
Crime in Neighbourhoods of New York City using Spatial Data Analysis.
2021 1st International Conference on Artificial Intelligence and Data
Analytics (CAIDA). Riyadh.
doi:https://doi.org/10.1109/CAIDA51941.2021.9425120
Alqahtani, A., Garima, A., & Alaiad, A. (2019). Crime Analysis in Chicago City. 10th
International Conference on Information and Communication Systems
(ICICS). Irbid, Jordan. doi:https://doi.org/10.1109/IACS.2019.8809142
Andresen, M. A., & Malleson, N. (2015, July 2). Intra-Week Spatial-Temporal
Patterns of Crime. Crime Science, 4(12). doi:https://doi.org/10.1186/s40163-
015-0024-7
Anser, M., Yousaf, Z., Nassani, A. A., Alotaibi, S. M., & Kabbani, A. (2020).
Dynamic linkages between poverty, inequality, crime, and social expenditures
in a panel of 16 countries: two-step GMM estimates. Economic Structures,
43. doi:https://doi.org/10.1186/s40008-020-00220-6
Aycardo, R. A., & Tubay, J. M. (2017). Crime Hot Spot Detection: A Case on Illegal
Drug. Retrieved from
https://www.academia.edu/39908468/Crime_Hot_Spot_Detection_A_Case_o
n_Illegal_Drug_Trafficking_in_Manila_National_Capital_Region
Barrera, D., Cagang, S., & Capistrano, D. (2013). Spatial and Temporal Maps of
Reported Crimes in Dumaguete City, Negros Oriental, Philippines. PRISM,
18(1), 11-24. Retrieved from
https://d1wqtxts1xzle7.cloudfront.net/37777100/SPATIAL_AND_TEMPOR
AL_MAPS_OF_REPORTED_CRIMES_IN_DUMAGUETE_CITY__NEGR
OS_ORIENTAL__PHILIPPINES-libre.pdf?1432951115=&response-content-
disposition=inline%3B+filename%3DSPATIAL_AND_TEMPORAL_MAPS
_OF_REPORTED_CR.pd
Bendler, J., Ratku, A., & Neumann, D. (2014). Crime Mapping through Geo-Spatial
Social Media Activity. Building a Better World thorugh IS. International
Conference in Information System (ICIS).
doi:https://core.ac.uk/download/pdf/301363221.pdf
54
Braga, A. A., Brandon, T., Papachristos, A. V., & Hureau, D. M. (2019, September
08). Hot spots policing of small geographic areas effects on crime. Campbell
Systematic Reviews, 15(3), 1-88. doi:https://doi.org/10.1002/cl2.1046
Chainey, S., Tompson, L., & Uhlig, S. (2008). The Utility of Hotspot Mapping for
Predicting Spatial Patterns of Crime. Security Journal, 21(1 and 2), 4-28.
doi:https://doi.org/10.1057/palgrave.sj.8350066
Cheng, W., Rao, Y., Yang, J., Chen, Y., Peng, L., & Hao, J. (2022, August 31).
Identifying the Spatio-Temporal Characteristics of Crime in Liangshan
Prefecture, China. International Journal of Environmental and Public Health.
doi:https://doi.org/10.3390/ijerph191710862
Crime hotspots. (2023, March 25). Retrieved from Wikipedia:
https://en.wikipedia.org/wiki/Crime_hotspots
Crime Mapping. (2023, 05 09). Retrieved from Criminal Justice: https://criminal-
justice.iresearchnet.com/criminology-research-methods/crime-mapping/
De Melo, S. N., Pereira, D. V., Andresen, M. A., & Matias, L. F. (2017, April 12).
Spatial/Temporal Variations of Crime: A Routine Activity Theory
Perspective. International Journal of Offender Therapy and Comparative
Criminology, 62(7). doi:https://doi.org/10.1177/0306624X17703654
Gonzales, A. R., Schofield, R. B., & Hart, S. V. (2005). Mapping Crime:
Understanding Hot Spots. Office of Justice Programs. Washington, D.C.:
National Criminal Justice Reference Service. Retrieved from
https://nij.ojp.gov/library/publications/mapping-crime-understanding-hot-
spots
Gramlich, J. (2020, November 20). What the data says (and doesn’t say) about crime
in the United States. Retrieved from Pew Research Center:
https://www.pewresearch.org/short-reads/2020/11/20/facts-about-crime-in-
the-u-s/
Gu, D., Newman, G., Park, Y., Lee, J., & Kim, J. (2019, April). Neighborhood
decline and mixed land uses: Mitigating housing abandonment in shrinking
cities. Land Use Policy, 505-511.
doi:https://doi.org/10.1016/j.landusepol.2019.02.033
He, Z., Lai, R., Wang, Z., Liu, H., & Deng, M. (2022). Comparative Study of
Approaches for Detecting Crime Hotspots with Considering Concentration
and Shape Characteristics. International Journal of Environmental Research
and Public Health, 19(21), 14350.
doi:https://doi.org/10.3390/ijerph192114350
55
Huafang, X., Liu, L., & Yue, H. (2022). Modeling the Effect of Streetscape
Environment on Crime Using Street View Images and Interpretable Machine-
Learning Technique. International Journal of Environmental Research and
Public Health, 19(21). doi:https://doi.org/10.3390/ijerph192113833
Kitchen, P. (2007). Exploring the Link between Crime and Socio-Economic Status in
Ottawa and Saskatoon: A Small-Area Geographical Analysis. Research and
Statistics Division . Department of Justice Canada. Retrieved from
https://www.justice.gc.ca/eng/rp-pr/csj-sjc/crime/rr06_6/rr06_6.pdf
Lauritsen, J. L., & White, N. (2014). Seasonal Patterns in Criminal Victimization
Trends. Bureau of Justice Statistics, U.S. Department of Justice. Washington,
D.C.: Office of Justice Programs. Retrieved from
https://bjs.ojp.gov/content/pub/pdf/spcvt.pdf
Loeber, R., & Farrington, D. P. (Eds.). (1998). Serious and violent juvenile offenders:
Risk factors and successful interventions. Sage Publications.
Malleson, N., & Andresen, M. A. (2016). Exploring the Impact of Ambient
Population Measures on London Crime Hotspots. Journal of Criminal
Justice, 46, 52-63. doi:https://doi.org/10.1016/j.jcrimjus.2016.03.002
Malleson, N., & Andresen, M. A. (n.d.). Using Social Media Data to Assess Spatial
Crime Hotspots. Retrieved from
https://d1wqtxts1xzle7.cloudfront.net/53104062/Using_Social_Media_Data_t
o_Assess_Spatial_Crime_Hotspots-libre.pdf?1494606480=&response-
content-
disposition=inline%3B+filename%3DUsing_Social_Media_Data_to_Assess_
Spatia.pdf&Expires=1686910234&Signature=fh4Bo
Mondal, S., Singh, D., & Kumar, R. (2022). Crime hotspot detection using statistical
and geospatial methods: a case study of Pune City, Maharashtra, India.
GeoJournal, 5287-5303. doi:https://doi.org/10.1007/s10708-022-10573-z
Neiss, M. (2015). Does Public Transit Affect Crime? The Addition of a Bus Line in
Cleveland. The Journal of Economics and Politics, 22(1). Retrieved from
https://collected.jcu.edu/jep/vol22/iss1/3
Newton, A. D. (2014). Crime on Public Transport. In G. Bruinsma, & D. Weisburd
(Eds.), Encyclopedia of Criminology and Criminal Justice. New York:
Springer. doi:https://doi.org/10.1007/978-1-4614-5690-2
Philippine National Police. (n.d.). Crime Information, Reporting, and Anaylysis
System (CIRAS). Retrieved 05 10, 2023, from
https://didm.pnp.gov.ph/images/james_vio/Crime_Information_Reporting_an
d_Analysis_System_CIRAS.pdf
56
Quezon City Local Government. (2023, April 2). Press Release. Retrieved from
https://quezoncity.gov.ph/: https://quezoncity.gov.ph/qcpd-reports-lower-
crime-rate-for-first-quarter-of-2023/
Ratcliffe, J. (2010). Crime Mapping: Spatial and Temporal Challenges. In Handbook
of Quantitative Criminology. doi:https://doi.org/10.1007/978-0-387-77650-
7_2
Sackett, C. (2023, May 03). Neighborhoods and Violent Crime. Retrieved from
Evidence Matters:
https://www.huduser.gov/portal/periodicals/em/summer16/highlight2.html
Saeed, R., & Abdulmohsin, H. (2023). A Study on Predicting Crime Rates through
Machine Learning and Data Mining using Text. Journal of Intelligent
Systems, 32(1). doi:https://doi.org/10.1515/jisys-2022-0223
Schaefer, R. T. (2008). Encyclopedia of Race, Ethnicity, and society (Vol. 1). Sage.
Shah, N., Nandish, B., & Manan, S. (2021). Crime forecasting: a machine learning
and computer vision approach to crime prediction and prevention.
doi:https://doi.org/10.1186/s42492-021-00075-z
Sypion-Dutkowska, N., & Leitner, M. (2017, March 8). Land Use Influencing the
Spatial Distribution of Urban Crime: A Case Study of Szczecin, Poland.
International Journal of Geo-Information, 6(3), 1-23. Retrieved from
https://www.mdpi.com/2220-9964/6/3/74
Taclay, K. D., & Taclay, R. J. (2022). Spatio-temporal analysis of crime-related
incidents in Bayombong, Nueva Vizcaya. AIP Conference Proceedings 2472,
050024 (2022). American Institute of Physics.
doi:https://doi.org/10.1063/5.0092708
Thomson, R., Espin, J., & Samuels-Jones, T. (2020). Green Crime Havens: A Spatial
Cluster Analysis of Environmental Crime. Social Science Quarterly.
doi:https://doi.org/10.1111/ssqu.12773
Wang, Y., Yu, W., Liu, S., & Young, S. D. (2019, March 29). The Relationship
Between Social Media Data and Crime Rates in the United States. Social
Media + Society, 5(1). doi:https://doi.org/10.1177/20563051198345
Weisburd, D. L., & McEwen, T. (2015, July 12). Introduction: Crime Mapping and
Crime Prevention. Retrieved from SSRN: https://ssrn.com/abstract=2629850
Wortley, R., & Townsley, M. (2017). Environmental Criminology and Crime
Analysis: Situating the Theory, Analytical Approach and Application. In R.
Wortley, & M. Townsley (Eds.), Environmental Criminology and Crime
Analysis (2nd ed.). London: Routledge.
57
Xiangyu, Z., & Tang, J. (2018). Crime in Urban Areas: A Data Mining Perspective.
Computers and Society, 20(1). Retrieved from
https://arxiv.org/abs/1804.08159
Yadav, N., Kumar, A., Bhatnagar, R., & Verma, V. (2019). City Crime Mapping
Using Machine Learning Techniques. In A. Hassanien, A. Azar, T. Gaber, R.
F. Bhatnagar, & M. Tolba, The International Conference on Advanced
Machine Learning Technologies and Applications (AMLTA2019) (pp. 656-
668). Springer, Cham. doi:DOI:10.1007/978-3-030-14118-9_65
Zhang, S., Sha, D., Dong, B., Ruan, S., Qiu, A., Li, Y., . . . Yang, C. (2020).
Spatiotemporal Patterns and Driving Factors on Crime Changing During
Black Lives Matter Protests. International Journal of Geo-Information, 9(11),
640. doi:https://doi.org/10.3390/ijgi9110640
58