GA-facilitated KNN classifier optimization with varying similarity measures

Abstract

Abstract-Genetic algorithms are powerful tools for knearest neighbors classifier optimization. While traditional knn classification techniques typically employ Euclidian distance to assess pattern similarity, other measures may also be utilized. Previous research demonstrates that GAs can improve predictive accuracy by searching for optimal feature weights and offsets for a cosine similarity-based knn classifier. GA-selected weights determine the classification relevance of each feature, while offsets provide alternative points of reference when assessing angular similarity. Such optimized classifiers perform competitively with other contemporary classification techniques.This paper explores the effectiveness of GA weight and offset optimization for knowledge discovery using knn classifiers with varying similarity measures. Using Euclidian distance, cosine similarity, and Pearson correlation, untrained classifiers are compared with weightoptimized classifiers for several datasets. Simultaneous weight and offset optimization experiments are also performed for cosine similarity and Pearson correlation. This type of optimization represents a novel technique for maximizing Pearson correlation-based knn performance. While unoptimized cosine and Pearson classifiers often perform worse than their Euclidian counterparts, optimized cosine and Pearson classifiers typically show equivalent or improved performance over optimized Euclidian classifiers. In some cases, offset optimization provides further improvement for knn classifiers employing cosine similarity or Pearson correlation.

Application of wrapper approach and composite classifier to the stock trend prediction

“…For instance, it is simple to understand and interpret, and it is able to handle nominal and categorical data and perform well with large data set in a short time. In this work, we use C4.5 decision tree to predict the direction change of stock price because C4.5 decision tree performs well in prediction application as report in Peterson et al (2005).…”

“…The kth nearest neighbor (KNN) algorithm (Kelly et al, 1991;Peterson, Doom, & Raymer, 2005) is a classification algorithm based on closest training example feature space. The training phase of the algorithm consists of storing the feature vectors and class labels of the training samples.…”

KB-CB-N classification: Towards unsupervised approach for supervised learning

“…Other similarity measures have been applied individually in the decision rule for classification purpose. For example density similarity measure used in density based classification [16] and data gravitation based classification [17].The classification method proposed in [18] with varying similarity measures (Euclidean distance, cosine similarity, and Pearson correlation) represents the first attempt. However, combining between other similarity measures like density and gravity has the potential to present a better view of the data distribution and hidden patterns within the training samples.…”

Method for increasing the computation speed of an unsupervised learning approach for data clustering

“…In order to analyze such data sets, clustering methods are often used as an integral tool for data preprocessing [1]. This is especially the case when data is recorded from multiple sources in an uncontrolled environment.…”

“…An objective of clustering is to identify parts of the data that has high degrees of similarity with other parts of the data, and group the similar parts together into clusters. Similarity can be measured by many means such as Euclidean distance, Manhalobis distance, cosine similarity, and Pearson correlation [1].…”