Documentation.txt

Data Processing:
	
	Missing values in dataset were handle by assigning a value of .000001. Instead of 0, this value
	will prevent the total probablity from being 0.

	Continuous values are treated with a probability density function. All continuous data are assumed to have
	a continuous distribution. Average and Variance are calculate of each classifier and its respective attributes.
	It is then plugged in to a function getContinuoutP, which outputs the probability of that value.

	getContinuousP is the probabilty density function.

	public double getContinuousP(String value, double average, double variance) {
        double x = Double.parseDouble(value);

        double term1 = Math.sqrt(2 * Math.PI * variance);
        term1 = 1 / term1;

        double term2 = (x - average);
        term2 = Math.pow(term2, 2);
        term2 = (term2 / variance) * (-.5);
        term2 = Math.exp(term2);

        return term1 * term2;
    }

Evaluation:

	1. Accuracy: 0.7594065792302731
	2. Accuracy: 0.7623091808213287
	3. Accuracy: 0.7597290905181682
	4. Accuracy: 0.7566114814018491
	5. Accuracy: 0.7624166845839604
	Five cross validation accuracy (average): 0.7600946033111159