A Machine-Learning Based IoT Smart Home System to Detect and Reduce Urban Insecurity in Uganda: A Case of Kampala Metropolitan Area

This study sought to develop and test a machine learning based prototype of a smart home security system for Kampala Metropolitan area in Uganda. The researchers used a Design Science Research (DSR) method to execute the project. The researchers identified a problem- which was high level of insecurity and crimes around homes in Kampala and its metropolitan areas, defined objectives, collected requirements, designed and developed the artifact, then demonstrated, evaluated, and communicated the artifact to stakeholders. The team created a working prototype of a home-security system that was later tested, demonstrated, validated among potential users. Practically, this project was implemented in a simulated environment using Tinkercad Arduino software. The developed system prototype was simulated on screen and tested on at least one home to ensure its effectiveness in reducing unauthorized entry, robbery, and other forms of crimes that are widely prevalent in Kampala today. One limitation of the project was that Arduino board used was quite limited on the number of pins that were available for use. The initial project idea had featured a temperature sensor to that was intended to regulate the temperature in the home by making use of the fan. However, the researchers did not implement this due to the limited pins and the capacity of the Arduino Uno R3 board that was used. Higher capacity industrial tools could provide a more realistic design of a similar system. The project simulated an environment of a smart home security system that can be used to detect and reduce crime. This study is part of a wider project to test ML and IoT systems in building secure smart homes. The authors envisage the application on more advanced ML algorithms and techniques in improving the system.