Paper Title
Automated Waste Segregation System using Trained Optical and Material Sensors with User Communication Capabilities

Abstract
As the world’s population grows, so does the amount of generated waste which continues to be an issue being faced around the world. Waste segregation is an effective way to lessen waste that could go to landfills while increasing the amount of recyclable materials. In this paper, a segregation system was developed using trained optical and material sensors to categorize the waste, while a mechanical segregating system was introduced. Aside from this, a web application where users will be able to view data gathered and will be able to validate the system’s categorization of wastes. By using this system, waste may be categorized into four: metal cans, plastic bottles, paper and other wastes. A model was trained using a dataset to recognize these four types of waste, while an inductive sensor aids in recognizing the material. The mechanical system consists of servo motors that rotate flaps which allow the trash to fall into the target receptacle. The current model has an accuracy of 83.54% but can be improved using the web application, where users will validate images captured by the system to improve the machine learning model. Keywords - Solid Waste Trash, Segregation, Machine Learning, Neural Networks.