Twitter is one of the biggest information hubs of our century and things being so, a lot of information is posted and consumed within a given period of time .One of the topics that have caused a real buzz over the past few years has been the out of control sense of insecurity from school shootings to muggings and even robbery. And that is the reason why I decided to settle on the topic. Over the past few months we have witnessed an unprecedented amount of insecurity and this owes a number of reasons. Some may be social / economic which are linked to poverty and some are as a result of the high rate of joblessness and inequality among different people.
By analyzing these trends in various parameters we will be able to analyze the various aspects of it and potentially give a likely prediction on what to expect in future. This will help in future planning and also help curb the menace. By analyzing the social influence and getting the relationship between the various parameters we can easily get the required parameters we need to alter the changes necessary to make it a safer place. By employing predictive analytic we can visualize the relation between some of the crime hot spots and the relation to the various social-economic parameters, it is a crucial aspect especially when it comes to long-term planning and making considerable legislative decisions regarding a locality.
By getting a visual display of the time of day a certain crime takes place and the place it does, we can use this data to predict the trends as well as the situations that leads to such an occurrence, thereby helping us take control of the situation. The advantage behind visual aids we can easily analyze and it makes the entire exercise a lot more effective.
When it comes to working with the various parameters linked to a certain ordeal or a continued menace by critically analyzing it a lot of aspects a revealed as opposed to other means.
1. ANALYSIS " PREDICTION
1.1 DATASET
The dataset we will be analyzing is from California and it gives a good detailed analysis of the various cities and the various aspects that come with each city as we shall see. There are a good number of indicators that will give us a great idea of the situation on the ground. The given data set will also be very crucial in painting an image of the situation on the ground.
From an overall population of 33,691,769 people.
Figure 2.1 A chart showing the aggravated assault against property crime.
The table above was obtained from the reported crimes in the state of California in the period of 2013, and cuts across 467 cities; it gives a rough estimate of just how prevalent the crime is in the cities.
After that, I went ahead to analyze the relationship between the cities with the highest aggravated assault and the lowest one in a bid to check the reason behind such a stack deviation.
Figure 2.2
The bar graph above shows the relation between violent crimes and a given population.
Due to its rough nature we have decided to look at a smaller set of data in relation to this.
Figure 2.3 A figure of violent crime in relation to population.
As we can see above from the smaller data set of the cities lined up from the most densely populated areas, the crime rate generally is higher and directly proportional to a given area with a few deviations. These deviations can be attributed to an areas level of education, culture and in some cases the social status. By analyzing it we find that
Figure 2.3 chart showing relation between population and various crimes.
2. Preprocessing
It is at this stage of the analysis of data that the collected information is analyzed, and most of the other features get to be filtered out for the sake of analysis. This stage is where the dataset is put into frameworks which will enable further visualization so as to give a more detailed output and one that can easily be examined and from which inferences can be made. The benefit of applying R programming linguistic which is also an extension that is very operational in a statistical analysis. The other advantage of using R is that it helps in coming up with visual displays as well as creating. Statistical computing and graphics are heavily applied in this computing and this aspect
With an ever-increasing popularity, R has great functionality and the ease of integration with python makes it so much easier to use, this forms a great background for which soft wares are used. One thing I noted while using is its static graphics, they are capable of producing high-quality static graphs. Since it is an interpreted language once the software is installed it can easily be accessed through the command-line-interpreter, which has an interface similar to that of a “Matlab”
Software. The fact that it supports matrix arithmetic and another wide array of structures such as vectors, arrays and data frames. R also has the capability of supporting procedural programming which entails among others.
In our case, we will be using R language to analyze data packages that are linked to the obtained data. This means we will be applying R libraries with a variety of statistical linear and non-linear models. This means the models can be manipulated along various parameters in order to give various aspects previously unavailable on other platforms. Take for instance a parameter such as the time within which crime tends to occur, one will find great similarities when being analyzed and it is from these similarities we get a good understanding of the nature of the crimes and also the locations that these crimes take place.
Figure. 2.5.
3.2 Exploratory Analysis
In the above figure we can clearly see a direct relation from obtained data where I have gone ahead and sorted it in descending orders of population. According to this data collected the nature of violent crimes has connections to various community aspects. And from the data the time it takes place and the people most affected are more or less very similar. We can also see that violent crimes are also directly related to the high number of robberies taking place. This means that the same characters who take place in the violent robberies will likely have an influence with the other phases.
On further insight one is likely to get on more research and analysis is the link between joblessness, drug usage and crime rate. When young people are jobless that will result to a lot of free time which will result to engaging in experimental activities which may lead to a lot of drug exposure in a bid to fit in to the masses or as a means to escape the sense of despair that is associated with joblessness and no sense of hope. And with this comes drugs, though not always directly linked but in most studies . one will likely unearth the stark reality that a there is a link between the jobless nature of these young people to the high crime rate. And with the high crime rate so is the case of organized gangs, which tend to give a sense of belonging to their members.
When it comes to dealing with such problems, social media may not always give the best solutions despite having been the source of my initial research. This is because it has a lot of unregulated information filled with people’s opinions which are not always sound and tend to satisfy people’s egos and fulfill their own gains. One of the best ways to deal with such a situation, which has also been highly suggested has been to take tougher measures in gun control. In as much as it is a right to be a licensed gun holder in such developed countries such regulations need to be reviewed, this is because when they infringe the rights of innocent people and lead to reckless homicides then that is just downright unacceptable. And above all else, it goes against all provisions of human rights.
Back to the applications of data analytics, when one is coming up with a data set to use in the process of researching using R programming language some consideration more so when it comes to the object-oriented aspect of it entails will be the type of data set and the type of output one will require. As discussed earlier one of the advantages of coming up with datasets organized r programming platform is the ease of integrating graphical visuals. This means that be it object-oriented programming or analysis as long as one has the ability to use Python then a lot can be achieved within a short span of time and at the same time give a good level of accuracy.
Another field in which the programming language r is applied in happens to be machine learning. It is a perfect tool when it comes to teaching neural networks as well as analyzing the rate of linear regression, which means that the process can be fast-tracked. But that is just a deviation from the main topic at hand.
So in the final analysis in as much as we want to give certain aspects to a given research carrying out an intensive detailed analysis based on credible tools and a certified data set is the first and perhaps the most crucial part of the entire process. It is also important to put in mind that intensive research and R programming is not exclusively meant for statisticians and data scientists but for anyone willing to learn.
