This section illustrates various works from literature that are carried out in order to identify traits of character for violence detection. The works are categorized based on factors which have been used for detecting violence. This literature forms the basis for selection of data for modeling stage. The data which is taken comprises of both i.e., data from literature and data from cloud sources. Now let us understand literature as follows:
In this article, we focused on work done by prominent researchers and different case studies of Ghana [22], Police Service of Northern Ireland Statistics and Research Agency (NISRA) [23], Ethiopia [24], Reports published by World Health Organization (WHO) [25], Fordham University (New York) [26], and Jordan Institute for families [27] to identify the factors resulting in violent behaviorand took cloud data [19] to maintain the confidentiality, integrity and availability of an organization. Violence is motivated by several factors. Some of the factors are highly visible and much cited by the researchers termed as determinate or known or certain factors, while some are less cited but having larger effects on the criminal behavior are named as indeterminate or unknown factors or uncertain factors throughout this research. Mapping these determinate and indeterminate factors can help policymakers, professionals and practitioners to identify criminal behaviors and build a surveillance system that may help in violence detection. Now let us understand each of them one by one.
Ebenezer S. OwusuAdjah et al. [22] found in their study that married women from Ghana reported factors that increase the possibility of creating places for domestic violence. Authors analyzed& gathered risk factors by using procedure of forward selection and multivariate logistic model on information provided by Ghana Demographic and Health Survey (GDHS) in 2008. At last, they found in their data that 1525 ever-hitched lady, and 33.7% had at any point experienced aggressive behavior at home. They observed that theaverage for confronting aggressive behavior at home was 35% for those ladies who live in metropolitan regions and the average of abusive behavior at home was 41% higher for ladies whose spouses at any point encountered their father beating their mother. So, living place, family history related to crime plays a greater role in determining the criminal behaviors among people which is less noticed by the researchers and hence finds a lower position while formulating the policies by the policymakers. Hence it is an indeterminate factor. Generally, it was noticed that ladies whose spouses drink too much alcohol face domestic violence 2.7 times higher than ladies whose spouses do not drink or do not take too much vine, so it is considered as determinate factor.
Michael A. Koenig et al. [28] found in their studies that violence (physical, sexual, and domestic) associated with individual factors of childlessness, economically not strong persons, some sort of pressure, and transmission of violence intergenerationally are important factors to violence. Hence, they termed as certain or determinate factors. Furthermore, the dowry system of India generates a relation with a higher risk of violence and level of dowry is highly determined by the financial condition of the family which in turn results in crime as pointed out by authors, hence it becomes an indeterminate factor leading to violence.
Sylvie Mrug et al. [29] investigated emotional desensitization which leads violence where external issues played role as a mediator of connection between exposure to violence into two categories like preadolescence and violent behavior in late adolescence. As we have seen some adolescents are exposed to violence in house, community, and college. Hence, they are termed as known, determinate, and certain factors. It has been represented in low levels of internalizing symptoms that high level violence has been linked to emotional desensitization. Even long consequences of desensitization are undefined. Hence on a broader note we take long term emotional of desensitization as an indeterminate factor which has a wider impact on criminal behavior.
As per Fordham University, New York [26], there are many factors that lead to aggressive behavior which contribute to violence, no one can stop by saying that this is the only definition of factors regarding aggressive behavior. General or known factors affected with aggressive behavior includes vandalism or fighting, excessive use of drug, alcoholic abuse, plans for violence, easy access to weapons but there are many factors that may be indeterminate that leads violence like extreme requirement for respect, feeling of low self-esteem, early misbehave with children, bullied, standing with violence in house, community, or in media.
Lyn Francis et al. [30] explored how ladies judge their experience of abusive behavior at home and also tried to ignore or end violence. There are many complex barriers that create pressure on women to keep silence on disclosing abuse, action to finish domestic violence. Some women did not acknowledge and even some women did not realize that their relationship was indefinite and sometimes tried to deny or minimize the maltreatment to adapt to the aggressive behavior at home. Authors found in their study that woman does not recognise nor even acknowledge abuse in their relationship, this took long time to provide best service. If they acknowledge violence at time then there will be some services on time which may help in minimising violence. Hence “Denying or minimizing the abuse by women” considers as indeterminate factor that leads violence.
Prothrow-Stith et al. [31] explored why manhandled kids experience issues in learning and studied the effect of experiencing childhood in vicious environmental elements and insecure neighbourhoods. Violent surroundings play an important role to grow with bad effects, that’s why it encourages more violence. So unsafe neighbourhood may lead to an indeterminate factor that leaves an important impact on violence.
Devadoss et al. [32] analyzedthe causes of youth violence. Both deadly and non-deadly assaults involving youth play widely to the weight of early demise, hurt, and infirmity. Basically, youth violence profoundly hurts not exclusively its causalities, yet in addition their families, companions, and networks. Authors worked on factors like Poor family functioning, Membership of a gang, Poverty in the community, Inequality, Influences of mass media but there are some indeterminate factors that should be considered like Dropping out of school, Denial of opportunities, Hiking of prices related to basic requirements of survival, Long drawn strike, Fake encounter, Run away from home, Violent video games, and Treatment by the teacher.
According to the Police Service of Ireland & NISRA (Northern Ireland Statistics and Research Agency) [23], any type of violence committed online or by Internet-based activities should be flagged. Authorities are involved in finding the reason of violence based on determinant factors like email, social media, websites, messaging platforms, etc., but they should go by the indeterminate factor like gaming platform or smart devices. Figure 3 shows the number and types of online crimes for Northern Ireland by the type of offenses.
Reva et al. [33] found in their study that targeted to examine and to expand the understanding of the working authorities who contribute to this area, to recommend possible ways to stop the imposed marriages. Constrained relationships are relationships where one of the two life partners does not consent to marriage or does not agree to the marriage. Brutality, dangers, or some other type of pressure is involved to realize the relationships and cause various exploitations. So, we can take forced marriages as an indeterminate factor that has an impact of violence happening.
World Report on Violence and Health [34], found that approximately 565 children, adolescents and young people into the age limit of 10 to 29 years die each day just because of interpersonal violence across the world. Mostly these cases may be seen from region of South-East Asia and rapid growth of violence escalation among youth in each Nation, culture and community is disturbing.
Kaufman et al. [24] found that violence based on gender is a genuine public concern in Ethiopia. As per author’s concern, harassment is known factor for gender-based violence in Ethiopia and there are other serious factors that lead violence as intimidation giving low social status to female students. These should be considered as unknown or indeterminate factors. WHO found in a study about health domestic violence related to women [35] that 71% of Ethiopian women faced physical or sexual brutality by their husband. This type of brutality/violence leaves a bad impact in their country, community and relations.
Pretus et al. [36] found an emotional element in his research that can affect any individual to get involved in violence among different groups for the values that are at stake. Sacred values are related to human emotions, for this people are ready to fight or even ready to die. People are always ready for the defence of these values, and they are very close to the people [37]. It does not matter what is the significance of them. These may be religious or secular like Holy Land, The Nation etc., these examined like critical to draw disputes recalcitrant [38,39,40]. Hence sacred values may be the indeterminate factor.
Jordan Institute for Families [27] released practice notes for predicting the violence and as usual they found known factors that leads violence like humiliation, access to weapons, experiencing childhood abuse or aggression in the home and found some indeterminate factors like boredom and a sense of powerlessness, feeling a sense of injustice or oppression.
Virginia Saez et al. [41] presented a state-of-the-art of research that addressed the mediatization of the phenomenon of violence in schools, from the mid-1970s to 2015. Their study showed the advances in research, predominantly Latin American, on the ways in which the media present the phenomenon of violence, its relationship with young people and the school. Authors categorised their work in three states: the contents and forms of violence in the discourses of the media information; informative coverage on young people and their connection with violence; and the school as a privileged setting for the mediatization of violence. Theoretical and methodological aspects involved in the research are examined. Subsequently, a review is made of the dimensions, violent episodes that still remain to be explored so it can be counted as indeterminate factor.
World Health Organization (WHO) [25] prepared data in terms of report on health and violence, found cause for death. As per the report, these death rates vary based on the Gross National Income (GNI). Death rates in low- and middle-income countries are more than twice as high (32.1 per 100,000) as those in high-income countries (14.1 per 100, 000). This can be seen in given figure that there is a huge amount of differences in death rates among WHO regions. WHO found in report that rates of Homicides are three times higher than rates of suicide in the regions of Africa and America as shown in Fig. 4. An average rate of suicide is higher than double rate of homicide in the regions of South-East-Asia and Europe. By these findings, we can take the income level of countries as an indeterminate factor.
Labrum Travis et al. [42] reviewed reasons of violence at home towards family members. Authors found about persons who have mental illness, involved in violence, generally they make trouble for caretakers, but not intentionally. As per author, family members faced violence, who have direct contact with mental illness person. Last year victimization was 20% higher. People living with dysfunctional behavior are at a humbly expanded chance of carrying out viciousness, lopsidedly prone for focus relatives which people really do commit ruthlessly. Authors found known factors behind the violence like recent victimization, mental health treatment, hostility, criticism in their work but there are indeterminate factors that authors talked about like nonadherence to medications and verbal aggression.
According to Rutgers [43] there are several contributing indeterminate factors in violence like penchant for narcissism to treating the casualty with no respect or regard, utilization of a fierce way of behaving, and ability to control the person in question and following. As per the study of Centres For Disease Control and Prevention under the chapter of violence prevention [44], violence related to youth made an understanding of factors that make public more vulnerable to victimization. Aggression in behavior, drug involvement, tobacco, alcohol, poor behavior with community, emotional distress, uncounted beliefs, bad attitudes are the risk factors that lead to youth violence but many violence factors that are too far from agencies they are just like unknown but have a significant role in violence, poor monitoring & supervision of children, parental substance abuse or criminality, and socially disorganized neighbourhoods. Violence is taking over humanity. To overcome violence, many prominent researchers did lots of work [45,46,47].
Chihlin et al. [48] performed work to overcome domestic violence by taking education as an indeterminate factor with the help of services of cloud computing. Authors considered gender paradigm and much involvement in technology as the factors leading to violence and presented novel method so that public can do legal work for their safety, education on family values, and family therapy supported by professional community members via cloud computing platform. Since cloud computing service can make education accessible to the needy people based on their demand.
Omar Sabri et al. [49] presented pros and cons of each factor and presented plan for picking best answer while selecting cloud computing service to overcome violence with cloud-based data. Authorsused two famous models of Information Systems Success, McLean and DeLone [50] model for accessing few elements that should consider by an organization when making the decision of adopting cloud computing services.
Diana Freed et al. [51] presented the role of mobile devices, and services of cloud computing playan important role in intimate partner violence (IPV). Theseinclude domestic abuse (determinate factor), stalking, and surveillance of victims by abusive partners. Authors conducted interviews with 40 IPV professionals and nine focus groups with 32 survivors of IPV and revealed a complex set of socio-technical challenges that stem from the intimate nature of the relationships involved and the complexities of managing shared social circles in IPV ecosystem, New York City. Both IPV professionals and survivors felt that they did not possess adequate expertise to be able to identify or cope with technology enabled IPV, and there are currently insufficient best practices to help them deal with abuse via technology. Many prominent researchers did the work regarding security challenges in cloud [52] and influencing factors in the decision to adopt cloud computing in the private sector [53]. Edeh et al. [54] worked on cloud security challenges by considering sacred nature of educational data, authors examined the implications of cloud security challenges on education and found that security [55] is very significant to the successful migration and implementation of cloud technology in the educational sector. It also shows that the growing security limitations associated with cloud computing technology has the tendency to discourage many educational institutions from adopting cloud services.
According to recent research by Palanivinayagam A. et al. [56], conventional crime detection and machine learning-based algorithms are unable to properly forecast crime trends because they are unable to produce essential prime qualities from the crime dataset. Their strategy is geared toward improving the topic machine learning algorithm’s precision by extracting the most salient attributes, such as time zones, crime probability, and crime hotspots, and doing vulnerability analysis. Concerns about security and privacy in relation to unmanned aerial vehicles were briefly analysed by Siddiqi M.A. et al. [57]. As a means of addressing these issues, the report also provided solutions and recommendations. In addition to providing a high-level overview of UAVs/drones, this paper updated the reader on the latest in relevant rules, classification, architecture, and communication protocols. The document also covered use cases, security flaws, countermeasures against various threats, and constraints. The report closed with a discussion of future research directions and some suggestions for bolstering the safety and privacy of UAVs.
Chen Zhi et al. [58] presented violence detection by Unmanned Aerial Vehicle (UAV) on the platform of cloud and used cameras to monitor public places, shoot videos and send them to the cloud platform to detect the violence in the videos, which can discover the violence in public places in time and give early warning to prevent criminal incidents. On the basis of the aforementioned works, many determinate and indeterminate factors that contribute to the environment of violence are identified. These works demonstrate not only how these factors contribute to violence, but also demonstrate how they might be included into future efforts to detect violence in real-time streaming media. Consequently, this research integrates these factors, as well as factors extracted from cloud data based on related works, in order to discover the most influential personality traits that lead to violent behavior. The motivation behind this work is need to address the issue of data imprecision when developing methods for violence identification in videos. This imprecision results from classifiers’ reliance on erroneous and incomplete data, which produces an inaccurate decision function. This also occurs when scores generated by many classifiers are fused; this is known as the imperfection problem. This research is also motivated by the belief that algorithms for detecting violence would perform better if they explicitly account for the indeterminacy, uncertainty, and imperfection of data at both the representational and decision levels. In fact, machine learning techniques are inseparable from indeterminacy and uncertainty as data that is used during the training phase is most of the time imprecise, incomplete, indeterminate, and noisy. The problem is also faced in classifier learning because the primary objective of classifier learning is prediction accuracy. Handling indeterminate and uncertain data should be a major concern of classifier learning since indeterminate data may impact not only interpretations of data and designed models but also impacts the sole purpose of the classifier, that is, prediction accuracy. Moreover, the generalization beyond that data, the process of induction, is still affected with uncertainty. This research, which focuses on the study and development of neutrosophic techniques for better handling of uncertain and indeterminate data in designing algorithms, is driven by the growing need for the development of effective and sophisticated techniques for handling data imperfections while designing approaches for violence detection.