This paper focuses on short-term prediction of Gold price trends, which is a crucial area for investors. The proposed model a Novel Distance Weighted K-Nearest Neighbors with Feature Extraction Algorithm (NDWKNN-FEA) approach that integrates KNN with a probabilistic method derived from Bayes’ theorem. This approach addresses the problem of assumptions made by distance functions in KNN classification, and considers both centric and non-centric data points in the computations of probabilities for target instances. The proposed model is compared with standard classifiers including LDA, PCA, RNN and ICA, and the results show that NDWKNN-FEA outperforms the others. The paper presents a state-of-art prediction model for Gold price trend prediction that can be used by researchers and investors from both financial and technical domains.

Gold Price Prediction, K-Nearest Neighbors, Bayes ‘Theorem, Naive Bayes, Probabilistic Method;