3. Literature Review:
In our paper we used three ways to get the leaf classification, one that was Random Forest, the other being Naïve Bayes and the third being Keras which is also known as Neural Networks. There have been many other cases like ours that have used Random Forest. Nantian Huang, Guobo Lu and Dianguo Xu have been one of the many who have used Random Forest and succeeded. In their paper, “A permutation importance based feature selection method for short term electricity load forecasting using Random Forest,” they talk about how they got load forecast through predictions. Load Forecast, is the planning, operating and scheduling of traditional power networks. The random forest in this was used to find the performance of each featured …show more content…
Random forest is used in this paper but usually to predict via using univariate method. But this time they used a multivariate method that being Random Forest. Random forest is good for microarray data and shows when the predictive data is “noisy, it can be used when the number of variables is much larger than the number of observations.” Their results in this paper showed that random forest has comparable performance compared to others including SVM and KNN. It also tells us that the gene results shows us small sets of genes thus a small predictive accuracy. In our project we are using high ranges of accuracy, but showing small set of accuracy just proves that under immense pressure random forest works well and gives out the best accuracy possible. This overview of the article gave us a good reason to chose Random …show more content…
Before using this technique, they were getting an accuracy of 0.807 and they could not get any other method to increase their accuracy. After the usage of neural network their accuracy increased to 0.837. It also states that their accuracy was better but at the same time this was the fastest method, they state that this was “much quicker method to deploy new fruits.” Therefore, stating that neural networks is a faster way of getting accuracy and more of an accurate measure and thus it would be perfect for our
In our paper we used three ways to get the leaf classification, one that was Random Forest, the other being Naïve Bayes and the third being Keras which is also known as Neural Networks. There have been many other cases like ours that have used Random Forest. Nantian Huang, Guobo Lu and Dianguo Xu have been one of the many who have used Random Forest and succeeded. In their paper, “A permutation importance based feature selection method for short term electricity load forecasting using Random Forest,” they talk about how they got load forecast through predictions. Load Forecast, is the planning, operating and scheduling of traditional power networks. The random forest in this was used to find the performance of each featured …show more content…
Random forest is used in this paper but usually to predict via using univariate method. But this time they used a multivariate method that being Random Forest. Random forest is good for microarray data and shows when the predictive data is “noisy, it can be used when the number of variables is much larger than the number of observations.” Their results in this paper showed that random forest has comparable performance compared to others including SVM and KNN. It also tells us that the gene results shows us small sets of genes thus a small predictive accuracy. In our project we are using high ranges of accuracy, but showing small set of accuracy just proves that under immense pressure random forest works well and gives out the best accuracy possible. This overview of the article gave us a good reason to chose Random …show more content…
Before using this technique, they were getting an accuracy of 0.807 and they could not get any other method to increase their accuracy. After the usage of neural network their accuracy increased to 0.837. It also states that their accuracy was better but at the same time this was the fastest method, they state that this was “much quicker method to deploy new fruits.” Therefore, stating that neural networks is a faster way of getting accuracy and more of an accurate measure and thus it would be perfect for our