QUESTION

Phase 2 Methodology to compare the old and new systems and the systems analysis. The other attachment is phase 1 and the related document you need to complete for this assignment. ( the PDF is part of this paper that you need to use as per the instruction.)

SOLOTION

1.0 Methodology to Compare the Old and New Systems and Systems Analysis

1.1 Introduction

The methodology employed in designing the new system should include the new concept in software development identified as Machine Learning. Machine learning is a subfield of artificial intelligence that mainly refers to a machine's ability to imitate a human's intelligent behavior. The technology has been used in handling complex cases where humans cannot perform perfectly.

Machine learning would now prove critical in designing the new violence detection system. With the adoption of Machine learning in the project, experts indicate that the system would now be fully efficient, as it would now be able to solve the big elephant in the room (Perry, 2013). According to software engineers, a machine learning system primarily uses the information acquired from its surrounding environment to solve its current issues. Developing the new violence prediction system using machine learning would now prove critical in solving the surge in crime reported by clients regularly.

The new violence prediction system would primarily have machine learning as its main engine. Machine learning would now prove instrumental in helping the new system predict cases of violence that would occur in the area, thus giving warning to the tenants. As per software developers, the new system would analyze data from the internet, mainly social media, thus proving vital in helping the system determine the current trend in crime (Perry, 2013). The system would also be able to identify the movement of terror groups that have caused havoc to residents in the region.

The violence prediction system would also be granted access to the criminal database and the nation's database on citizens, thus enabling it to access data on particular individuals in the country or region (Perry, 2013). The process would now prove critical to the system, as it would enable the system to monitor criminals constantly, thus giving a warning to residents of the premises of any impending danger.

2.0 The methods that will be used to compare the old and new systems

The old and the new systems primarily have lots of differences, thus making it easy for them to be compared by developers and researchers. The first metric that can be used to compare the two systems is security or safety. According to reports, the previous Airbnb violence detection system was inferior. Reports indicate that the systems could not protect Airbnb dwellers from attacks because they could not warn them of any pending danger.

 Reports also indicate that the old systems could not prevent the cyberbullying of the company's clients, making it another weak link in the old systems (Peixoto et al., 2021). However, the current new system has significantly improved in this area. For instance, the current is built through machine learning technology, thus making it perform its duties perfectly. According to reports, current technology makes it possible to detect crime or violence and send a warning before it occurs. The machine learning technology the current system uses plays a vital role in violence prediction as it learns new techniques daily, thus making detecting and reporting crime before it occurs simple.

Another metric that can be used to compare the two models is the efficiency and effectiveness of the systems. Reports indicate that the old system was far from efficient or effective. For instance, the inability of the old system to detect and give warnings to individuals and the company proves how ineffective it is, as it cannot solve the only problem assigned to it.

3.0 Literature Review

According to Mahmoodi and Salajeghe's (2019) article, "A classification method based on the optical flow of violence detection," the authors indicate that violence detection is one of the most common and challenging topics in intelligent video surveillance systems. The authors indicate that there has been a growing demand for violence detection systems as individuals have proven beyond doubt that the technology would prove vital in helping reduce the surge in crime rate witnessed in the past (Garca-Gómez et al., 2016). According to Mahmoodi & Salajeghe, 2019, recent technological advancement has therefore proved vital in developing new and better violence detection systems that would help protect individuals from attacks. The authors indicate that the introduction of machine learning in the development of violence detection systems has proved instrumental in developing modern systems that now make it easy for individuals to be safe (Mahmoodi & Salajeghe, 2019). Since machine learning systems are known to self-develop, individuals are now sure that the system will constantly update itself on the current threats and also identify possible methods that can be employed to limit or prevent the attacks from occurring.

According to Febin et al., 2020, in their article, "Violence detection in videos for an intelligent surveillance system using MoBSIFT and movement filtering algorithms," action recognition is an active area of research in computer vision that has attracted significant attention from different scholars globally. The idea of detecting violence and crime has been dramatically welcomed among computer scientists around the globe, thus proving vital in solving the crime menace (Febin et al., 2020). Additionally, the authors indicate that the development of an intelligent system for surveillance would also prove critical in solving the crime rate in a region. The authors also indicate the importance of machine learning in violence detection. As per the authors, machine learning has helped solve the crime issue in several regions globally, efficiently predicting crime and warning the relevant authorities. In the case of Airbnb, adopting the violence system developed by the machine learning process would now prove critical in solving crime in the region.

4.0 Systems analysis diagrams

Old system

Current system

5.0 Conclusion

The old Airbnb violence detection methods, functioned flawlessly for a long time; however, new developments have revealed that the Airbnb system has faults. The accuracy of the previous methods is essentially low because they seldom provide accurate predictions on situations that have yet to occur at the premises. The new Airbnb system would now prove critical in solving the issue of the surge in crime in the region. The new violence detection system would now prove accurate since it is based on a machine learning model. The machine learning model would now prove critical in the violence detection system, as it would primarily enable the system to have a better chance at detecting crime since it can learn new techniques that are used to launch attacks from time to time. As a result, the new violence detection systems would prove to be more productive compared to the old model, as it will critically gather data and analyze it, thus providing the right prediction, which would therefore prove vital in helping the Airbnb company and its clients stay safe at all times.

6.0 References

Abdali, A. M. R., & Al-Tuma, R. F. (2019, March). Robust real-time violence detection in video using cnn and lstm. In 2019 2nd Scientific Conference of Computer Sciences (SCCS) (pp. 104-108). IEEE

Febin, I. P., Jayasree, K., & Joy, P. T. (2020). Violence detection in videos for an intelligent surveillance system using MoBSIFT and movement filtering algorithm. Pattern Analysis and Applications, 23(2), 611-623.

García-Gómez, J., Bautista-Durán, M., Gil-Pita, R., Mohino-Herranz, I., & Rosa-Zurera, M. (2016). Violence detection in real environments for smart cities. In Ubiquitous computing and ambient intelligence (pp. 482-494). Springer, Cham.

Mahmoodi, J., & Salajeghe, A. (2019). A classification method based on optical flow for violence detection. Expert systems with applications, 127, 121-127.

Makadia, H. (2022, October 3). No code/low code vs. traditional development - which team should you pick? Maruti Techlabs. Retrieved November 19, 2022, from https://marutitech.com/no-code-low-code-vs-traditional-development/

Peixoto, B. M., Lavi, B., Dias, Z., & Rocha, A. (2021). Harnessing high-level concepts and visual and auditory features for violence detection in videos. Journal of Visual Communication and Image Representation, 78, 103174.

Perry, C. (2013). Machine learning and conflict prediction: a use case. Stability: International Journal of Security and Development, 2(3), 56.

.