SVM, Neural Network ABSTRACT
In the previous couple of years, the quantity of patients owning liver issue has expanded too much. By the ongoing investigations, liver illnesses are getting to be a standout amongst the most fatal infections in a few nations. To diminish the hazard in liver illnesses, identification of the disease at the prime stage and concluding the sickness is imperative to treat a patient in better manner and decrease the risk successfully. In HIT Industries(Heath Care IT), use of computer based tech is increasing in a fast and subtle manner, as it becomes less hazardous to identify the diseases in an automated way. Starting late, basic undertakings are made to expand the utilization of computer based technology in the different regions of diagnostic process, since mistakes in therapeutic analytic frameworks can result in truly deceptive medicinal medications. Simulated intelligence is crucial in Computer controlled diagnosis. To predict the disease in the early stage according to the subtle symptoms is very difficult for the medical researchers. To process the data from the voluminous medicinal database, Data mining is the most utilized innovations in Health Care Industries. AI methodologies can be utilized on different liver patient datasets to foresee an exact and definite diagnosis and analysis. In the case of identifying liver problems, ML (machine learning) and AI (artificial intelligence) increases the exactness of the prediction and diagnosis. It makes step by step and worthy calculations. It is also useful to deliver the measurements, contiguous information and progressive examination on the dataset. The aim of this project is to increase the exactness of the accuracy of the diagnosis of liver disease at an early or preliminary stage. The main objective of the project is to classify the liver patients from the healthy ones.
2.1 INTRODUCTION
Dataset Specification: This dataset was downloaded from the Kaggle Website https://www.kaggle.com/uciml/indian-liver-patient-records .
This Dataset is taken from University Of California (UCI) ML(Machine Learning) Repository [http://archive.ics.uci.edu/ml].. In 2013, Lichman M organized the dataset. The data is set up on the survey taken on the locality of NE (North East) region of Andhra Pradesh.
The dataset has 11 columns. The total number of rows and columns are followed by Number of Male groups – 441 Number of Female groups – 142 Total Number of rows - 583 The last column Dataset is used to classify the patients into two groups. ( patient with Liver disease or no disease)
Requirements: Anaconda Navigator Jupyter Notebook
Packages: Pandas,Numpy,Matplotlib,seaborn,sklearn,Keras