GEOVISUALIZATION ANALYSIS OF GENDER-BASED VIOLENCE IN MEXICO CITY USING 3D MAPPING APPROACH

Abstract

The study of criminal acts is a topic of great interest to society and with the increase of violence against women in the world makes working with this type of data an important subject, where it can be used to detect areas with high crime rates to help in decision making. This work uses open data sets available from government entities of Mexico City, where three sectors of interest were studied: economic, education, and transportation around the criminal investigation reports made by the government. For the processing and analysis of spatial data, QGIS was used with the help of open-source libraries, the 2D and 3D modelling was carried out to create heatmaps and detect hotspots. It was found that the Historic center of Mexico City is the most insecure and that the Cuauhtémoc district stands out as the most dangerous zone, for transport stations, the ones that had the most intersections with other stations are the ones that presented greater problems of gender violence, in the economic sector there was a high incidence in the limits of each district, contrary to the results obtained on the educational services.

Full text

GEOVISUALIZATION ANALYSIS OF GENDER-BASED VIOLENCE IN MEXICO CITY
USING 3D MAPPING APPROACH
M. Saldana-Perez 1, C. Palma 1, *, Y. Z. Contreras 1, N. Carrillo 1, M. Moreno-Ibarra 1
1 Instituto Politécnico Nacional (IPN), Centro de Investigación en Computación (CIC), Av. Juan de Dios Batiz, s/n, 07320, Mexico
City, Mexico - (amagdasaldana, cpalmap2020, ycontrerasj2020, ncarrillog2020, marcomoreno)@cic.ipn.mx
Commission IV, WG IV/9
KEY WORDS: Crime analysis, Crimes against women, GIS, Heatmap, Open data, Open source.
ABSTRACT:
The study of criminal acts is a topic of great interest to society and with the increase of violence against women in the world makes
working with this type of data an important subject, where it can be used to detect areas with high crime rates to help in decision
making. This work uses open data sets available from government entities of Mexico City, where three sectors of interest were
studied: economic, education, and transportation around the criminal investigation reports made by the government. For the
processing and analysis of spatial data, QGIS was used with the help of open-source libraries, the 2D and 3D modelling was carried
out to create heatmaps and detect hotspots. It was found that the Historic center of Mexico City is the most insecure and that the
Cuauhtémoc district stands out as the most dangerous zone, for transport stations, the ones that had the most intersections with other
stations are the ones that presented greater problems of gender violence, in the economic sector there was a high incidence in the
limits of each district, contrary to the results obtained on the educational services.
* Corresponding author
1. INTRODUCTION
In Mexico, violence and crimes are relevant subjects given the
fact that just in 2020 were reported 27.6 million crimes into
several categories (INEGI, 2021). One of these categories, and
the one this paper is based on, is gender crimes. Gender crimes
can be defined as all crimes of violence against women just for
the fact that they are women (Sanchez, 2020). Another
definition is all the crimes and aggressions, including physical
violence and psychological violence that are committed against
the women, several manifestations could be considered violence
like sexual, economic, and social violence (Torres, 2007).
One of the most dangerous cities in Mexico is Mexico City. In
the 2020 Encuesta Nacional de Victimización y Percepción
sobre Seguridad Pública (ENVIPE) was reported the prevalence
of a crime with a rate of 35,238 for every 100,000 women. For
that reason, Mexico City is where more gender crimes are
reported (ENVIPE, 2020).
The Mexican government provides a repository with open data,
where this data is available to anyone who needs it. Open data is
categorized into 12 general sectors related to the city, for
example, health, economy, education, infrastructure, security,
justice, etc. In the portal of Mexico City, there exists data of
investigations about gender crimes reported since 2016. This
data is georeferenced data, and for that reason, it is possible to
do an analysis using a Geographic Information System (GIS).
The geographic information system (GIS) are systems or tools
computed-based that are used to collect, store, process, analyze
and visualize geographic information in an efficient way. The
geographic information Science is the study of the science that
investigates geographic information, and seeks ways to
represent phenomena in the real world. (Longley et al., 2015).
These systems allow making geographic representations or
map-view representations, where these representations provide
a realistic view of the events that occurred.
In this case, we propose the use of free software tools such as
QGIS for doing the data analysis, modeling and
geovisualization with the aim that this study can later be
replicated.
This paper presents the study of violent acts made against
women obtain from the public repository of Mexico City in a
period of two years from 2019 to 2020 around three sectors of
interest, such as: economic (convenience stores), transportation
(subway and subway bus) and education (private and public
schools) in Mexico City. The analysis applied implements 3D
geospatial analysis techniques to see the incidence of gender-
based violence, in order to help the population and
organizations in decision-making to prevent crime.
This work is made up of five sections. The first section is the
introduction, the second one presents the related works. In the
third section, we propose the methodology, which considers
both 2D and 3D geo-visualization. The fourth section shows the
obtained results and finally in the last section conclusions are
made.
2. RELATED WORK
In the following section, some studies related to the research
proposal are described. The works are related to 3D mapping,
crimes against women, and geo-visualization in 2D and 3D.
Since the incidence of crime is a topic of interest to the
population, different approaches have been developed, which
include the implementation of 3D mapping techniques. In a
study conducted by the German city of Cologne (Wolff &
Asche, 2009), the authors introduce 3D geovisualization
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume X-4/W2-2022
17th 3D GeoInfo Conference, 19–21 October 2022, Sydney, Australia
This contribution has been peer-reviewed. The double-blind peer-review was conducted on the basis of the full paper.
https://doi.org/10.5194/isprs-annals-X-4-W2-2022-241-2022 | © Author(s) 2022. CC BY 4.0 License.
241

techniques to produce crime maps in order to identify hotspots
using a kernel density estimation (KDE). They create a GIS
plug-in that calculates the minimum distance between a building
and the closest crime that has occurred. Likewise, Hashim et al.
(2019) applied 3D visualization to identify criminal patterns of
urban crime (13 types) through hotspots by applying Emerging
Hot Spot Analysis (Space-Time), they also applied Ordinary
Least Squares (OLS) Regression to determine the factors that
influence the patterns, the study used ArcGIS Pro 2.4 tool to
perform the analysis.
Crime incidence can be seen using heatmaps to identify areas
prone to violence against women, Garfias Royo et al. (2020)
present a proposal that shows heatmaps derived from surveys
carried out in the Corregidora in Querétaro with the purpose of
locating places susceptible to violence against women and
determine the causes of the incidence of gender-based violence
in this area of Mexico.
The proposed method is described as low cost since it collects
data through surveys carried out in the place of study with the
aim to help the prevention of gender crime. The result of this
work was a geospatial analysis of areas of violence in Querétaro
City and concludes that there is a relation between social norms,
gender structure (such as appearance, sexuality, capabilities, and
religion), and gender violence. It mentions the importance of
generating georeferenced data relating to gender violence since
that information is important to decrease the problem.
Gender violence is not a problem that happens just in Mexico as
seen in Crime against women in Chandigarh: A GIS analysis
(Bhattacharyya, 2016). A study about crime against women was
done with the help of surveys of gender violence victims. The
surveys were applied in some cities such as Agartala, Kohima,
Imphal, Shillong and Guwahati. The purpose of this work is to
find the common factors of violence and use the results to make
better decisions. The authors identified that some of the
elements that determine gender violence are: being a woman,
belonging to a certain religion, age, having a disability, being
from another country, and speaking a different language.
Furthermore, it was found that some of the places where women
felt less safe are: public transport, streets, bus stops, and taxis.
In previous investigations, the importance of specifying the
geographical areas with the highest incidence of gender
violence is established, in the same way, Kahlon (2014) carry
out a GIS analysis in order to identify the most susceptible
spaces and reasons for gender crimes. This study makes use of
the records of eleven police stations in the city of Chandigarh in
India, a geospatial and spatio-temporal analysis of gender
crimes was developed, resulting in that the highest incidence of
gender crimes occurred in informal neighborhoods and in areas
of student influx, which suggests that this criminal incidence is
highly related to the socioeconomic and demographic profile of
the locality.
Moreover, the use of open data is crucial for geospatial analysis,
in the following works the use of open data and its importance
are mentioned. In the study of Belesiotis et al. (2018) they
describe the use of crowdsourced and open data as crucial for
geospatial analysis, they used several heterogeneous open data
sources in order to create predictions of spatial distributions of
crimes and identify hotspots of crimes. Likewise, Groff and La
Vigne (2002) use an open data set made by the Vancouver
Police Department to predict crime. They develop a statistical
model to make the predictions (break and entries for residential
and commercial locations.), whereby with the help of mobile
GIS predictive maps were created in order to assist patrol units
to make decisions.
Similarly, a geospatial crime analysis study that employs the
United Kingdom crime statistics and analyses crime trends was
conducted to develop a web-based system that helps visualized
heatmaps and is capable of displaying crime trends in the UK
(Bonatsos et al, 2013).
Finally, a study that assesses the walkability of women based on
location and use of open data in New York City was performed
by Gorrini et al. (2021), this analysis concentrates on the
security issues that women face, the objective was to recognize
the most insecure geographical areas.
3. METHODOLOGY
The present work proposes a way to analyze data from different
open sources in a social, economic, and educational context to
face gender-based violence. The methodology takes into
consideration 2D and 3D modeling with the aim of showing in a
visual and realistic way the results obtained. Figure 1, shows the
four main steps of the methodology: data acquisition, data
preprocessing, geoprocessing, and geovisualization.
Each stage contains a series of steps to follow, for example, the
first stage, which consists of data acquisition, refers to
identifying the important data for the analysis to be carried out.
For its part, the preprocessing stage applies data cleaning, and
identification of relevant variables, among others. The third
stage has two sub-stages: two-dimensional and three-
dimensional modeling. Lastly, stage four presents the
visualization of the maps.
3.1 Data selection
The data used was obtained from different open data sources,
the gender crime investigations from Fiscalia General de
Justicia (FGJ) of the portal Abierto de Datos de la Ciudad de
México. Due to the fact that this data set incorporates records
from 2016, only those that occurred in 2019 and 2020 were
selected, since they use the same classification method and
reflect the current situation, in this period 496,863 reports were
registered (Datos abiertos CDMX, 2021).
The data crime repository has all the investigations of the
prosecution that occurred in the period between 2019 - 2020,
because of this a subset that contains only gender crimes was
generated. First, we created a subset with all the crimes
committed against women but, because not all violent acts
against women are considered gender-based crimes they were
separated according to their classification (Ramos, D., 2020). A
total of 65,343 reports were set apart.
The second dataset used in this work was collected from the
Directorio Estadistico Nacional de Unidades Economicas
(DENUE). From this set, two subsets were selected: economic
activities (convenience stores) and one focused on education
that includes any educational service, either public or private
(INEGI, 2022).
For the subsets of the economic and educational sectors, the
codes of the activities that belong to each one were identified.
The DENUE implements the North American Industry
Classification System (NAICS), when obtaining the code of the
sector the necessary data can be separated, for the economic
sector code 46 was used, due to the large number of business
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume X-4/W2-2022
17th 3D GeoInfo Conference, 19–21 October 2022, Sydney, Australia
This contribution has been peer-reviewed. The double-blind peer-review was conducted on the basis of the full paper.
https://doi.org/10.5194/isprs-annals-X-4-W2-2022-241-2022 | © Author(s) 2022. CC BY 4.0 License.
242

that are registered in the DENUE it was necessary to use a small
portion of the data, for this purpose the subsector retail and
minisupers with code 462112 was selected obtaining 3,135
records, and for the educational sector with code 61, a total of
11,249 schools were detected.
The third dataset contains the 12 subway lines and 195 subway
stations of the transportation of the collective metro system of
Mexico City, this dataset was also obtained from Portal de
Datos Abiertos de la Ciudad de Mexico (Datos abiertos CDMX,
2021). Within the public transport system is the Public
Passenger Transport Corridor System of Mexico City, also
known as Metrobús, it has 7 lines and 283 stations. This set of
data is also found on the portal.
For the 3D visualization and modeling of the city, it was
necessary to incorporate information from the Sistema Abierto
de Informacion Geografica de la Ciudad de Mexico (SIGCMX),
which provides cadastral data and ground records (Datos
abiertos CDMX. 2020).
3.2 Data preprocessing
Given the fact that the data collected belongs to different
sources and each one has a different format, the datasets require
preprocessing, this includes the cleaning and integration of the
data.
In this step, we consider the most important attributes that are:
investigation folder identifier, date, classification and type of
crime, district, latitude and longitude. Because the files handle
similar attributes such as the date of issue and start date, one of
these were dropped. Additionally, void and incomplete values
were removed.
3.3 Geoprocessing
To analyze and compile the geographic information the QGIS
software was used, which has an open source license, this
system was utilized because of its different functions and
available plugins that allow the application of multiple
algorithms to perform the required analyses useful for this
investigation. This phase has two main components: 2D and 3D
modeling. To create the modeling, the first layer a real-world
representation was added, the next one was the gender-based
crime layer, and lastly a layer of one of the three sectors:
transportation, economic or educational.
3.3.1 2D Geospatial crime model
As a first step two-dimensional geoprocessing was carried out
since the layers obtained during this process are later necessary
to conduct the 3D modeling. The first layer that is added is the
real-world representation, then a delimitation layer for the
districts in Mexico City, next the crime layer, and from the
previously selected sectors (transport, economic and
educational) areas of influence (buffer) are delimited to study
the incidence of gender crimes that have occurred.
The processing tools to perform crime analysis and modeling
that happened in Mexico City are part of the plugins available in
QGIS, first the data acquired from FGJ is loaded, then the
appropriate sector of interest is identified: educational centers,
economic stores, stations, and lines of the collective transport
system. Taking into consideration the selected points from the
area of interest we made a 300 meters (about 984.25 ft) buffer
for the first two sectors and a 1-kilometer buffer for
transportation (Figure 2).
Figure 2. 2D analysis and model.
In the second step, the count point algorithm is applied to the
buffer created in the last step to determine how many crimes
occurred around a place (in the buffer area). This point counting
is later used to apply a categorization with a color ramp, where
pink represents the lowest value and purple represents the
highest (number of crimes).
Figure 1. Proposed methodology for the analysis of different open data sources implementing map modelling to perform
geovisualization.
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume X-4/W2-2022
17th 3D GeoInfo Conference, 19–21 October 2022, Sydney, Australia
This contribution has been peer-reviewed. The double-blind peer-review was conducted on the basis of the full paper.
https://doi.org/10.5194/isprs-annals-X-4-W2-2022-241-2022 | © Author(s) 2022. CC BY 4.0 License.
243

In a similar way, the analysis carried out on the crimes that
occurred around a building was carried out, firstly with the
centroids algorithm, the midpoint of each construction were
identified, then a buffer of 50 meters was created from the
centroid obtained, like the analysis carried out for the sectors of
interest, the crimes committed in the delimited area were
counted, then the Join attributes by field process was applied,
later this value will be employed to identify the number of
crimes that occurred around a place along with the use of
identifying colors according to the quantity.
The heatmap was obtained from the heatmap algorithm (kernel
density estimation) the values used were radius: 0.005, 2000
rows, and 1750 columns. At the end of the process, it is
necessary to prepare the layer for 3D modeling in such a way
that the raster layer is converted into a vectorial layer with the
polygonizer algorithm (raster to vector) in order to implement
the height variable according to the density of crimes when
using 3D visualization (for the general gender-based crime
map).
3.3.2 3D Geospatial crime model
The 3D implementation allows volumetric information to be
added, such as mountains, buildings, routes, and elevations. In
this particular case, we can use the number of crimes as height.
One of the most important advantages of representing the
information in this type of cartography is the intuitive
symbology it offers since it facilitates the representation and
compression of the results.
In the same way as the 2D modeling, 3D modelling needs a
surface layer, the Google Map Terrain Hybrid was used to get a
digital elevation model (DEM) which is the base layer. This
version contains both geographical relief as well as the streets
and points of interest, this is important for the representation of
crime zone incidence, since this form allows the consultation of
various results of a specific area.
Subsequently, a three-dimensional environment of Mexico City
is created with a dataset obtained from SIGDMX, to determine
the areas and buildings affected by gender violence. The dataset
contains 2,373,869 registers which have an attribute
denominated level range, with this variable it was possible to
appoint the height for each building and form the model of the
city (Figure 3).
Figure 3. 3D model of Mexico City.
For the study of criminal cases in the transportation sector,
arbitrary height was given to the subway and subway bus routes
so that when the geovisualization stage was carried out, they
would stand out more and the routes could be seen better on the
map. The official colors of the Mexican subway were assigned
to distinguish between each transport line.
The results derived from the two-dimensional analysis done in
the 2D modeling step were used to create the 3D modeling
patterns of the crimes against the woman, with the help of
Qgis2threejs plugging.
3.4 Geovisualization
For the implementation, integration, and visualization of the
data crimes in a three-dimensional format, we use the
Qgis2threejs plugin which allows us to visualize the study area
depending on the sector. With this plugin, through the
generation of files, it is possible to see the maps in a web
browser and it can also support file formats like gITF. In this
stage the layers made in the previous steps are used to build
interactive maps where the public can consult the results
obtained from the 3D modeling, in a quick and easy way, they
can identify hotspots, and thereby identify the areas with more
crimes in Mexico City.
4. RESULTS
As a part of the results, heatmaps were generated where the
hotspots of gender crimes can be seen. The created maps
contemplate both an overview map of Mexico City and certain
areas of Mexico City to show specific cases in more detail.
In Figure 4, a general heatmap of Mexico City is presented,
since this overview uses all the data as a base, it is later used to
compare with the ones obtained from sectors of interest. It is
shown that the district where most gender crimes happened is
Cuauhtemoc as it has cases between 150-180 and 120-150. The
next district is Coyoacan followed by Gustavo A. Madero and
Venustiano Carranza.
The general view shown in Figure 4, uses a similar analysis
presented in Figure 5 since it makes use of all the records of the
crimes that occurred in a delimited range but does not apply
extra analysis by sector.
Figure 4. Gender crime heatmap of Mexico City.
Figure 5 shows the number of crimes that occurred near a
building within a radius of 50 meters, even though most of the
buildings are in a range of 0-10 crimes, there are areas with
ranges that can reach 50 crimes.
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume X-4/W2-2022
17th 3D GeoInfo Conference, 19–21 October 2022, Sydney, Australia
This contribution has been peer-reviewed. The double-blind peer-review was conducted on the basis of the full paper.
https://doi.org/10.5194/isprs-annals-X-4-W2-2022-241-2022 | © Author(s) 2022. CC BY 4.0 License.
244

Figure 5. 3D model hotspots using the Mexico City model.
The results obtained from the study of gender crimes in the
different sectors of interest can be seen in the next maps. Figure
6 corresponds to educational services that include both public
and private schools that range from elementary to universities,
trades schools, and others (music, art, among others), similarly
to the presented 3D heatmaps, it is shown that the districts
Cuauhtemoc and Coyoacan have the highest crime index, where
is detect that more crimes occur in the center of each district. To
identify the most dangerous places a radius of 300 meters was
implemented (the width of the cylinders on the map), the height
of the cylinders represents the number of crimes that occurred.
Although there are few educational services that have had more
than 120 criminal reports around them, a pattern can be noticed,
where when a peak appears on the map (represents the place
with the most cases), around this there are other peaks, although
smaller between the 60-100 crimes, which indicates that the
area around these schools is unsafe.
Figure 6. Hotspots of gender crimes occurring around
educational services.
The result from the same analysis, but in the economic sector,
that is, convenience stores throughout Mexico City are
presented in Figure 7, as can be seen in the heatmap, it can be
identified that the district with the highest crime index is
Cuauhtemoc, it was also found that in most cases there is a
higher crime rate in the boundaries of the city districts.
Figure 7. Hotspots of gender crimes occurring around the
economic sector.
Finally, in the result of the 3D visualization of the gender
crimes in Mexico City around subway routes, the analysis
presents that there exists a higher criminal incidence when there
is a greater number of subway station transfers, and the influx is
higher, likewise for the downtown area of Mexico City (Figure
8).
Figure 8. Heatmap of gender-based crimes on subway lines in
CDMX.
In Figure 9 the central zone of Mexico City, as well as some of
the subway stations near the Historic center of Mexico City, are
presented. It is evident that in this zone exists a higher index of
gender violence, to be more specific, in the station called
Guerrero, which has 2 intersections one with line 3 and another
one with line B, there were between 150-200 crimes. At the
intersection of the stations Balderas and Niños Heroes there
were around 100-150 crimes.
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume X-4/W2-2022
17th 3D GeoInfo Conference, 19–21 October 2022, Sydney, Australia
This contribution has been peer-reviewed. The double-blind peer-review was conducted on the basis of the full paper.
https://doi.org/10.5194/isprs-annals-X-4-W2-2022-241-2022 | © Author(s) 2022. CC BY 4.0 License.
245

In the north zone of the city, the index of crime rate is 0-50 (an
example of this is the Politecnico station). Moreover, in
Chabacano station where there is an intersection of 3 stations,
there are between 100-150 reports. These stations are lines 3, 8,
and 9. It can also be seen that, on the west side of the city, the
last stations of lines 3, 9, 6, and 7 (Universidad, Tacubaya y El
Rosario) stand out compared to previous stations, they have
between 50-100 crimes compared to 0-50 reported crimes
presented in the nearby stations.
Figure 9. Heatmap of gender-based crimes on subway lines in
CDMX.
Figure 10 shows the crimes committed around the subway bus
routes, a buffer of 1 kilometer was used. In a similar way to the
results presented above, the pattern that expresses the
concentration of gender crimes happening in the Historic center
of Mexico City is repeated. The stations that have intersections
with other subway bus routes also presented more insecurity.
Figure 10. Heatmap of gender-based crimes on subway bus
lines in CDMX.
A detailed view of the downtown area of CDMX that shows as
mentioned before a more detailed map presents the specific
stations that have the greatest problems. The crimes in subway
bus routes between Garibaldi in line 7 and Guerrero in line 1
have about 150-200 gender crimes, (Figure 11).
Figure 11. Heatmap of gender-based crimes on subwaybus
lines in CDMX.
5. CONCLUSION AND DISCUSSION
The use of both 2D and 3D geovisualization techniques allowed
the creation of heatmaps to help specifically detect points or
trends that occur in the data. By applying the study in different
sectors (economic, educational, and transportation), the need to
generate knowledge from the open data obtained can be
verified.
The areas with the highest crime rate were successfully
detected, during the investigation, it was found that, although
the three sectors present the Cuauhtémoc district as the place
with the most reported cases, certain patterns change according
to the analysis of the crimes from a chosen sector, in the
economic sector crimes tend to occur in the territorial limits of
the districts, however for educational services the opposite
occurs since more crimes occur in the center, on the other hand
in the transport sector the results are very similar between the
metro and Metrobus due to their concentration in the part of the
Historic center of Mexico City.
The use of open-source tools and resources and free data
allowed us to find crime patterns in Mexico City from the 3D
geoprocessing and geovisualization, which can then be used for
decision-making to prevent crime. The hotspot detection helped
determine the areas with the highest crime rates, so the
government can use this type of analysis to allocate resources to
these areas in order to face and prevent gender-based crimes.
Likewise, having a geovisualization of the data allows us to
observe certain patterns of the events that have occurred, which
would allow the government to deploy strategies to minimize
such events.
6. FUTURE WORK
Although a relationship was found between the sectors of
interest and the number of crimes against women that occurred
according to the industry, other areas could be used to find other
types of relationships or implement a multi-criteria space
analysis that takes several sectors of interest simultaneously.
Likewise, applying a study using the population density of
women by area to identify if there is a correlation between the
number of reports that have occurred and the number of people
who live in a place could allow us to associate if more crimes
occur in a place due to its insecurity or if the higher amounts are
related to the number of inhabitants that each delegation has.
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume X-4/W2-2022
17th 3D GeoInfo Conference, 19–21 October 2022, Sydney, Australia
This contribution has been peer-reviewed. The double-blind peer-review was conducted on the basis of the full paper.
https://doi.org/10.5194/isprs-annals-X-4-W2-2022-241-2022 | © Author(s) 2022. CC BY 4.0 License.
246

Due to the fact that the data present date and time variables,
they can be later used to carry out a space-time analysis.
Finally, this study applied the use of the 30 types of criminal
acts against women, research according to the type of crime to
generate more specific results than those presented can be
conducted.
ACKNOWLEDGEMENTS
The work was done with partial support from the Mexican
Government through the grants 1083730, 1083728, 1084083 of
CONACYT. The authors also are grateful to Secretaría de
Investigación y Posgrado of the Instituto Politécnico Nacional,
through the projects 20221894, 20221469 and grants
B200432,B200500, B200501.
We would like to thank the reviewers of this work for taking the
time and effort to read this paper, we appreciate the feedback
and opinions made to improve the work.
REFERENCES
Belesiotis, A., Papadakis, G., & Skoutas, D. (2018). Analyzing
and Predicting Spatial Crime Distribution Using Crowdsourced
and Open Data. ACM Transactions on Spatial Algorithms and
Systems, 3(4), 1–31. https://doi.org/10.1145/3190345
Bhattacharyya, R. (2016). Street Violence against Women in
India: Mapping Prevention Strategies. Asian Social Work and
Policy Review, 10(3), 311–325.
https://doi.org/10.1111/aswp.12099
Bonatsos, A., Middleton, L., Melas, P., & Sabeur, Z. (2013).
Crime Open Data Aggregation and Management for the Design
of Safer Spaces in Urban Environments. IFIP Advances in
Information and Communication Technology, 311–320.
https://doi.org/10.1007/978-3-642-41151-9_30
Datos abiertos CDMX. (2020, June 12). SISTEMA ABIERTO
DE INFORMACIÓNGEOGRÁFICA DE LA CIUDAD DE
MÉXICO. Retrieved February 1, 2022. From the website:
https://sig.cdmx.gob.mx/datos/
Datos abiertos CDMX. (2021, January 12) Portal de Datos
Abiertos Ciudad de Mexico: Víctimas en Carpetas de
Investigación. Retrieved February 1, 2022. From the website:
https://datos.cdmx.gob.mx/dataset/victimas-en-carpetas-de-
investigacion-fgj/resource/d543a7b1-f8cb-439f-8a5c-
e56c5479eeb5
Datos abiertos CDMX. (2021, January 12). Portal de Datos
Abiertos Ciudad de Mexico: Ubicación de líneas y estaciones
del Sistema de Transporte Colectivo Metro. Retrieved February
1, 2022. From the website:
https://datos.cdmx.gob.mx/dataset/lineas-y-estaciones-del-metro
Datos abiertos CDMX. (2021, January 12). Portal de Datos
Abiertos Ciudad de Mexico: Ubicación de líneas y estaciones
del Metrobús. (2022), Retrieved February 1, 2022. From the
website: https://datos.cdmx.gob.mx/dataset/geolocalizacion-
metrobus
ENVIPE. (2020, December 10). Encuesta Nacional de
Victimización y Percepción sobre Seguridad Publica. Instituto
Nacional de Estadística y Geografía. Retrieved March 10, 2022.
From the website:
https://www.inegi.org.mx/contenidos/programas/envipe/2020/d
oc/envipe2020_presentacion_nacional.pdf
Garfias Royo, M., Parikh, P., & Belur, J. (2020). Using heat
maps to identify areas prone to violence against women in the
public sphere. Crime Science, 9(1), 1-15.
https://doi.org/10.1186/s40163-020-00125-6
Gorrini, A., Presicce, D., Choubassi, R., & Sener, I. N. (2021).
Assessing the Level of Walkability for Women Using GIS and
Location-based Open Data: The Case of New York City.
Findings. https://doi.org/10.32866/001c.30794
Groff, E. R., & La Vigne, N. G. (2002). Forecasting the future
of predictive crime mapping. Crime Prevention Studies, 13, 29-
58.
Hashim, H., Wan Mohd, W. M. N., Sadek, E. S. S. M., &
Dimyati, K. M. (2019). MODELING URBAN CRIME
PATTERNS USING SPATIAL SPACE TIME AND
REGRESSION ANALYSIS. The International Archives of the
Photogrammetry, Remote Sensing and Spatial Information
Sciences, XLII-4/W16, 247–254. https://doi.org/10.5194/isprs-
archives-xlii-4-w16-247-2019
INEGI. (2021, September 22). ENCUESTA NACIONAL DE
VICTIMIZACIÓN Y PERCEPCIÓN SOBRE SEGURIDAD
PÚBLICA (ENVIPE) 2021 [Press release].
https://www.inegi.org.mx/contenidos/saladeprensa/boletines/20
21/EstSegPub/envipe2021.pdf
INEGI. (2022). Instituto Nacional de Estadistica y Geografia,
Retrieved January 25, 2022. From the website:
https://www.inegi.org.mx/app/descarga/default.html
Kahlon, S. (2014). Crime against women in Chandigarh: A GIS
analysis. International Journal of Management and Social
Sciences Research (IJMSSR), 3(9), 82-87
Longley, P. A., Goodchild, M. F., Maguire, D. J., & Rhind, D.
W. (2015). Geographic information systems and science (4th
ed.). Wiley.
Ramos, D. (2020, September 10). Ficha Metodologica Mujeres
victimas de delitos en carpetas de investigación de la fiscalía
general de Justicia de la ciudad de México. Retrieved March 21,
2022. From the website:
https://genero.congresocdmx.gob.mx/wp-
content/uploads/2020/10/2-CELIG-Ficha-Metodo-CarpetasInv-
2020.09.10.pdf
Sánchez De Los Monteros Arriaga, A. C. (2020). La violencia
de género en México, ¿en qué vamos?. Revista Digital
Universitaria, 21(4).
https://doi.org/10.22201/cuaieed.16076079e.2020.21.4.1
Torres, P. O. (2007, November). Análisis de la Clasificación
Mexicana de Delitos. Retrieved March 3, 2022, from:
https://www.corteidh.or.cr/tablas/26389.pdf
Wolff, M., & Asche, H. (2009). Towards Geovisual Analysis of
Crime Scenes – A 3D Crime Mapping Approach. In: Sester, M.,
Bernard, L., Paelke, V. (eds) Advances in GIScience. Lecture
Notes in Geoinformation and Cartography. Springer, Berlin,
Heidelberg. https://doi.org/10.1007/978-3-642-00318-9_22
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume X-4/W2-2022
17th 3D GeoInfo Conference, 19–21 October 2022, Sydney, Australia
This contribution has been peer-reviewed. The double-blind peer-review was conducted on the basis of the full paper.
https://doi.org/10.5194/isprs-annals-X-4-W2-2022-241-2022 | © Author(s) 2022. CC BY 4.0 License.
247