Sara Haghbayan; Behnam Tashayo; Mehdi Momeni
Abstract
Extended Abstract
Introduction
Today, one of the most complex issues in most countries is the high crime rate and the increase in social anomalies in them. One of these anomalies is residential burglary, which is one of the most widespread crimes in most countries of the world. Because spatial and ...
Read More
Extended Abstract
Introduction
Today, one of the most complex issues in most countries is the high crime rate and the increase in social anomalies in them. One of these anomalies is residential burglary, which is one of the most widespread crimes in most countries of the world. Because spatial and time play a very important and undeniable role in the formation of hot crime spots such as residential burglary therefore, by identifying the spatial and temporal of hot crime spots can be largely prevented. Previous studies have focused more on identifying and analyzing spatial crime hotspots and performing temporal analysis of crimes independently of spatial crime hotspots. However, in order to prevent the occurrence of these crimes in the future, a combination of time and spatial hot crime spots is needed to provide a more complete and accurate analysis. The aim of this study is to provide a systematic method for combining spatial and temporal information of residential burglary. The proposed method is based on spatial analysis and allows investigating the temporal distribution of events in hot crime spots. For this purpose, GIS capabilities have been used to perform statistical and graphical tests to identify and display crime hotspots. The results showed that hotspots follow a spatially clustered and temporally focused pattern. The research findings showed that the highest frequency of burglary is in hot spot No.4 in 2016 August, on Wednesday at 8 am, and the lowest frequency of burglary is in hot spot No.1 in 2018 January, on Sunday at 4 am.
Materials & Methods
The statistical tests used in this study include mean center, standard deviation ellipse test for clustering. The first step in identifying crime hotspots is to use the tests for clustering. For this purpose, in this study, the method of the average nearest neighbor is used. The results of residential burglary test for clustering showed that this crime is a cluster pattern in the study area. After proving to be clustered, graphical methods including point map display and kernel density have been used to display the hot crime spots. The results of the kernel density test cause to the identification and display of four spatial the hot crime spots in the study area.
The data used in this research include information on the time, place and type of crimes in the years 2015, 2016, 2017, 2018. The total number of crimes is 319073, of which 5573 were related to residential burglary, which was used as a statistical population in this study.
Results & Discussion
Statistical analysis was performed over a period of four years, which is equivalent to 48 months and 35064 around the clock for each hot crime spot. The results show that the highest incidence of crime in hot spot No.4 is equivalent to 1172 cases of residential burglary, which of all these four hot spot has a smaller area equivalent to 1117 hectares. Temporal analyzes of hot crime spots were performed annually, monthly, weekly and hourly. The results of the annual analysis of all four hot spots show that the highest rate of residential burglary is in 2016 and the lowest rate is in 2018.
The findings of this study show that the combination of spatial and temporal of hot crime spots analysis lump-sum by temporal analysis regardless of the spatial hot spots in monthly, daily and hourly intervals is significantly different. The combination of spatial and temporal of hot crime spots in the monthly interval shows that the maximum and minimum rates of residential burglary per month are different in these four hot spots. The highest number of residential burglary respectively occurred in hot spot No. 1 in October, in hot spot No. 2 in August, in hot spot No. 3 in June and in hot spot No. 4 in August. However, the results of the statistical analysis of time without considering the spatial hot crime spots show that August is the highest and April is the lowest. Daily statistical analysis shows that the highest number of residential burglary occurs in hot spot No. 1 and hot spot No. 3 on Friday, while in hot spot No. 2 it is Thursday and in hot spot No. 4 it is Wednesday. This analysis is different with a general daily analysis that shows Friday as the highest number of occurrences. Hourly analysis also shows that the peak of residential burglary in all four centers is at different hours; Thus, the peak of residential burglary areas in the study area is in the hot spot No. 1 hour 22, in the hot spot No. 2 hours 17, in hot spot No. 3 hours 12, in the hot spot No. 4 hours 8. However, statistical analysis of the time without considering the spatial hot spot shows the peak of residential burglary at 12 noon.
Conclusion
In this study, a new framework for the simultaneously displaying the pattern of crimes in two dimensions of spatial and time was presented, which can be used to identify the pattern of distribution of spatial and temporal of hot crime spots. The results of kernel density estimation analysis are four spatial-temporal crime hotspots where the spatial hotspot distribution pattern is clustered and the temporal of hot crime spots distribution pattern is focused. The results show that 78% of burglaries occur in these four crime hotspot, which cover only 25% of the total area of the study area. Therefore, by identifying the spatial and temporal of hot spots, crime can be largely prevented. This method is used to identify and display any type of crime in each study area and allows the identification and display of the combination of spatial and temporal hot crime spots.
Sara Haghbayan; Behnam Tashayo
Abstract
Extended Abstract Introduction Air pollution has become a life-threatening hazard with severe consequences. Previous studies have indicated that long-term exposure to air pollution can pose a significant threat to human health or even cause death. Usually, air quality is monitored by ground-based ...
Read More
Extended Abstract Introduction Air pollution has become a life-threatening hazard with severe consequences. Previous studies have indicated that long-term exposure to air pollution can pose a significant threat to human health or even cause death. Usually, air quality is monitored by ground-based stations that can collect data regarding temperature, humidity, pressure, and several pollutants such as Ozone (O3), Carbon Monoxide (CO), Carbon Dioxide (CO2), Sulfur dioxide (SO2), Nitrogen dioxide (NO2), and nanoparticles (e.g. PM1, PM2.5, and PM10). However, ground-based stations are costly, scattered, and often cannot cover large areas. These stations collect the concentration ofparticulate matter with a diameter of less than 2.5 µm (PM2.5) over a year.Collected data may be lost due to an unexpected shutdown of the device. Datacollected in ground-based stations are not sufficient by their own and as a result they are modeled. The resulting models also have flaws, so new resources are needed to solve this problem. One of these resources is the use of mobile sensors to produce high-resolution temporal and spatial air quality data. As opposed to traditional air quality monitoring stations, the use of dynamic and mobile sensors is quickly developing. These mobile sensors measure the concentration of the same air pollutants as those measured by ground stations. Land-use regression (LUR) models are increasingly used to estimate the level of PM2.5exposure in urban areas. Land-use regression models often use data received fromground-based stations. Therefore, modeling the concentrations of particulate matter in a city leads to a significant increase in modeling error. Data from mobile sensors can increase the accuracy of this contaminant modeling process. The present study aims to improve modeling accuracy by integrating ground-based stations with mobile sensors. Therefore, using the proposed framework, we can accurately estimate air quality at any time and place and provide higher resolution estimations for heterogeneous urban environments. Materials & Methods The study area covers Isfahan city. With a population of more than two million and an area of 200 square kilometers, Isfahan is located in central Iran. 13% of the total pollutants entering Isfahan belong to urban industries, 11% to domestic sources, and 76% of all pollutants belong to traffic related sources in Isfahan. Therefore, most of the PM2.5concentrations are generated by the transportation system in Isfahan. The effective solution to the air pollution problem needs to have a comprehensive understanding of the air pollution process. Such an understanding primarily depends on reliable records that can depict the temporal and spatial variations in air pollution which is not possible due to the limited number of ground-based stations. The proposed method of the present study is to combine ground-based stations with mobile sensors to increase the accuracy of PM2.5concentration estimation and modeling. One of the existing methods used to estimate PM2.5levels is land use regression. Previous studies used only ground-based stations to create this model, which was not sufficiently accurate. The present study sought to increase the accuracy of PM2.5concentration modelling in contamination values of near or beyond the threshold. Using the LUR model, a prediction map was generated usinga combination of ground-based stations and mobile sensor which helps us to reach a more accurateestimation and prediction of PM2.5concentrations in a heterogeneous region such as this city. Results & Discussion Reliable and accurate estimate of temporal/spatial distribution of air pollutant concentration cannot be achieved using a limited number of ground-based stations. The present study took advantage of 14 mobile sensors along with 7 ground-based stations. Results indicated that the root mean square error of the seven ground-based stationsequaled 1.80 while the RMSE of the combination of these stations equaled 0.59. The skewness index shows asymmetry of data as compared to the standard normal distribution.This index is used to determine whether the data distribution is normal or not. Skewnessvalue of standard normal curvesequals zero. In the histogram obtained from a combination of all stations, this value is 0.11, while in the histogram obtained from the ground-based stations, skewness value equals 0.8803. In general, the results indicated that integrating ground-based stations with mobile sensors results in a PM2.5concentration distribution which looks more like a normal distribution. The normality of data distribution implies that the histogram of data frequency is approximately a normal curve, and thus T-test is used to examine whether or not the results were significant. Conclusion In this study, a new framework was proposed to integrateground-basedstations and mobile sensors with the aim of improving the accuracy of PM2.5 pollutant concentration estimation. The results of the t-test show that with only ground-based stations, the actual pattern and its distribution over the city will fail. In fact, data received from mobilesensors provide additional data necessary for air pollution profiling.
behnam tashayo; Mohammad Sharif
Abstract
Extended Abstract Introduction With the developments in navigation, positioning, and tracking technologies, a large amount of moving point data (e.g., human, vehicle, animal) have been produced. Through moving an object in the course of time, a sequence of its position is recorded which is known as trajectory. ...
Read More
Extended Abstract Introduction With the developments in navigation, positioning, and tracking technologies, a large amount of moving point data (e.g., human, vehicle, animal) have been produced. Through moving an object in the course of time, a sequence of its position is recorded which is known as trajectory. Studying the behaviors of point objects and analyzing their trajectories have recently received great attentions among researchers in different fields of science, especially in geographic information science. Such studies contribute to better understanding of movement-behavior patterns of moving objects. Data mining, as one of the main approaches in geographic knowledge discovery, is normally used in moving databases to extract information from moving point objects’ trajectories. Analyzing the similarity of trajectories as one of the frequently used approaches in geographic data mining, is of great importance, which is normally performed by distance functions. The way an object moves is highly influenced from the condition and situation (known as context) where movement takes place. Therefore, merely analyzing trajectories from their spatiotemporal dimension may not clearly reflect the conditions and situations (contexts) during the move, and may lead to unrealistic results. Accordingly, most of the present distance functions are only able to measure the similarity of trajectories at spatial and spatiotemporal dimensions and neglect the underlying context during the move. Therefore, developing a new trajectory similarity measure approach, capable of handling a wide range of context information in conjunction with spatiotemporal dimension becomes crucial. Materials & Methods According to the complexity and variety in the moving objects’ datasets, especially in their context parameters, in this research, a fuzzy model is developed which is able to exploit a wide range of internal and external context information in similarity measure process of multi-length trajectories. Specifically, a fuzzy inference system (FIS) is used for measuring the similarity between trajectories based on qualitative and quantitative context information. To enhance the effectiveness of the system, a subjective hierarchical fuzzy inference system (HFIS) is designed and the membership functions and rules are generated and adjusted based on learning. The HFIS controls the rules by dividing a FIS into several simple sub-systems that are hierarchically connected to each other. The Multi-Objective Particle Swarm Optimization (MOPSO) algorithm is used for defining and adjusting the FIS. The performance of the system is assessed by applying it on real trajectories of pedestrians’ and cyclists’ datasets in the Central Park of Manhattan, New York, USA, while accounting contextual information. The movement parameters and context information were recorded in qualitative and quantitative formats. Two types of context information are used: 1) internal context (i.e., all of the properties that directly relate to the moving object, such as eagerness), and 2) external context (i.e., any factor that extrinsically influences the process of movement, such as slope). According to the nature of data, four models of spatial, spatiotemporal, spatiotemporal-internal context, and spatiotemporal-internal-external context are applied on the dataset to see the added value of each parameter. Results & Discussion The achievements of the implementations can be discussed from two main perspectives: 1) system performance, and 2) similarity results. From the system performance aspect, the control surface plots indicate the consistency of the rules in the developed fuzzy inference systems. In addition, the Pareto diagrams for both pedestrians’ and cyclists’ models cover the Pareto fronts in all four models in an appropriate manner. From the similarity results aspect that are shown in boxplots, the commonalities between both pedestrians’ and cyclists’ trajectories are very high at the spatial dimension in the first model, while by adding the temporal dimension as the second model, the similarity range at the spatiotemporal dimension increases. This issue is directly related to the traversed paths and the elapsed times of the moving objects. By importing the internal contexts (i.e., speed, direction, length, eagerness) as the third model, the similarity range again increases but the median decreases. However, the distribution in the pedestrians’ similarity values is much more than the cyclists’ values. The fourth model is composed of all the previous parameters plus the external contexts (i.e., path difficulty and slope). The same as previous trends, the similarity range increases but the median decreases. Although the number of trajectories that share common characteristics are low when taking into account all the spatial, temporal, internal contexts, and external contexts in the implementation, the results become much more realistic and close to reality for both datasets. In addition, to assess the influence of the internal and external contexts on the similarity output, a one-way analysis of variance (ANOVA) test, with a 95% confidence level, is applied on the results. Furthermore, a number of sensitivity analyses are applied on the datasets. The outcomes demonstrate meaningful influence of contexts on the similarity results of trajectories. Conclusion In this research a novel approach based on fuzzy systems is developed for multi-dimensional similarity measure of trajectories while accounting both internal and external context information. This approach enhances the previous similarity functions that only use spatial and temporal data. The developed models were applied on two real trajectory datasets, which were capable of handling both quantitative and qualitative information. The results indicate the effectiveness of the proposed context-aware similarity measure approach in revealing the commonalities between trajectories in multi dimensions.
