Missing values are prevalent in real-world datasets and they may reduce predictive performance of a learning algorithm. Dissolved Gas Analysis (DGA), one of the most deployable methods for detecting and predicting incipient faults in power transformers is one of the casualties. Thus, this paper proposes filling-in the missing values found in a DGA dataset using the k-nearest neighbor imputation...

k-Nearest Neighbor (k-NN) is one of the classification algorithms which becomes top 10 in data mining. k-NN simple and easy to apply. However, results are greatly influenced by scale input. All its attributes considered equally important Euclidean distance, but inappropriate with relevance each attribute. Thus, it makes decreased. Some more or less or, fact, irrelevant determining results. To o...

In case of insufficient data samples in highdimensional classification problems, sparse scatters of samples tend to have many ‘holes’—regions that have few or no nearby training samples from the class. When such regions lie close to inter-class boundaries, the nearest neighbors of a query may lie in the wrong class, thus leading to errors in the Nearest Neighbor classification rule. The K-local...

This research was made in order to see which method performance is better between the KNN (K-Nearest Neighbor) regression and multiple linear on Boston Housing data. The performace referred here MAE, RMSE, MAPE, R2. a predict something based closest training examples of an object. Meanwhile, forecasting technique involving more than one independent variable. comparison two methods results Mean ...