Comparative Analysis of Classifiers for Prediction of Epileptic Seizures

Epilepsy is a neurological disease in which people suffer from seizure attack and lose the normal function of brain. Almost 50 million people have epilepsy in the world due to which it has become the most common neurological disease. Early prediction of epilepsy helps patients to avoid epilepsy and live normal life. Many studies have been conducted for the early prediction of epilepsy. However, selection of the most appropriate classifier has always been a question that needs to be resolved. In this study, we are using six classifiers of machine learning which are KNN, Naïve Bayes, Linear Classification Model, Discriminant Analysis Model, Support Vector Machine and Decision Tree, to find the best classifier for the prediction of epileptic seizures, in term of accuracy. Dataset from “Kaggle” was used. Preprocessing and cross-validation of the data was carried out for training and testing of classifiers. The results depict that Naive Bayes classifier has a better average accuracy of 95.739% as compared to other classifiers. The future work of this study is to implement the suggested model in real time, so that the workload of medical members could be reduced.

Air Quality Monitoring System.

The project served to build a device that can measure the concentration of atmospheric pollutants such as PM2.5, PM10, O3, COx, NOx, and SOx. The purpose of doing so was to combat the fact that Pakistan lacks an AQI (Air Quality Index), and previous efforts have not been fruitful in this venture. The constant rise of lung diseases and high levels of atmospheric pollutants have made absolutely essential that Pakistan develop an AQI and atmospheric pollutant monitoring system. The project consisted of two parts; the background device that captures the concentration of gases and an online application that receives said concentrations and displays them visually such that it may be understood by the general public. This report focuses on the background device, its working and implementation. The device was initially designed to measure all the above mentioned parameters however, due to constraints in procurement locally available sensors were used that measure CH4 and CO in the atmosphere. A backup battery was installed so that the device remains operational even if the mains supply is powered off. The sensors are integrated with an ESP32 which is a microcontroller with a built-in WiFi that is used to send data across to the server through HTTP protocol. The data needs to be collected every 5 minutes and therefore the device needs to be able to store data on the chip until a WiFi connection is restored. The device is also robust against the harsh environments of Karachi, and reports any anomalies that it may find with the operation. The device although in full functionality misses a few important benchmarks. Firstly, the sensors currently used arent accurate enough, and secondly, the device only relies upon WiFi to send the data across to the server. This should be avoided in the next iteration of the design by using a GSM/GPRS module with a SIM card as theres no public WiFi available in Karachi