Aims to build a classification model that can provide a basic understanding of what types of personal characteristics are likely to cause stroke and the probability that those characteristics cause stroke.
The classification model can help doctors or researchers to predict the risk of a person under the stroke once been inputed with new data. We will also establish an unsupervised model, try to compress the dataset by identifying which features are most useful in distinguishing different examples. We can potentially discard the less significant ones for further analysis.
Though the model’s ability to provide accurate medical diagnosis may be somewhat limited, it provides a sufficiently adequate platform for self check-up. Based on the analysis, we would recommend those people who are in old age, with a higher BMI, suffering from heart disease and hypertension try to exercise more and be aware of the risk of getting a stroke in advance. In addition, people who work in the private sector and self-employees are also recommended for routinely taking relevant physical examinations.