To test the effectiveness of the random forest model, 33% of the data has been selected as the testing sample after shuffling the dataset. The random forest model first build on 67% of training dataset. The score for the training model was 0.633 whereas the score for the testing set was 0.601.  The scores of the random forest model demonstrate a moderate predicting power as the scores are relatively close to each other. Therefore, the model is possibly not behaving in a over predicting manner. 
        The sample predicting results are summarized in the follow table (Table 1) and a confusion matrix below demonstrates the testing result (Fig.3). The result shows that when conducting the prediction, the model predicted almost all the sample incidents results to be TRUE which is too extreme. The accuracy of the model is 62.01% while the precision of the model is 62% as well. The sensitivity of the model is  very high, 99.9% but the specificity is around 0.3%. With such high sensitivity, the model can be overestimating the possibility of food poisoning.