In the Philippines alone, the amount of citizens that rely on unsafe water for daily use and consumption is at an overwhelming amount due to the fast urbanization of the country and a lack of sanitation when it comes to several bodies of water. Previous researches have aimed their focuses at being able to identify several water sources and utilizing either a machine learning algorithm or an Internet of Things device coupled with the machine learning algorithm to help monitor and classify water supplies at their sources or at a water reserve and in identifying abnormalities that the researchers would need to monitor. Here, this research utilizes an I.o.T device coupled with a Naive Bayes algorithm to help classify the data inputted form a given sensor and a Convolutional Neural Network to help monitor images of a water sample than can identify both Microscopic and Macroscopic images from a given water supply. The effectiveness of bringing an Internet of Things device and combining the technology with Machine Learning Algorithms and making the system portable via an Android Application has given the system a profound portability and accessibility when compared to pre-existing water monitoring devices. The addition of image taking as well as geotagging has given the system more versatility in the field when it comes to wastewater treatment and water quality monitoring. The system created by the researchers was able to identify macroscopic images with an accuracy of 86.5% and microscopic images with 84% for the C.N.N. algorithm and 96.9% for the Naive Bayes algorithm. These diseases but also to help bring relief to those who may be unsure of their source of water. The researchers have decided to use a machine learning model, specifically a Convolutional Neural Network, to create a system that is able to detect impurities from an image of water fed into the system.