Progress in digital data acquisition and storage technology has resulted in exponential growth in high dimensional data. Removing redundant and irrelevant features from this highdimensional data helps in improving mining performance and comprehensibility and increasing learning accuracy. Feature selection and feature extraction techniques as a preprocessing step are used for reducing data dimensionality. This paper analyses some existing popular feature selection and feature extraction techniques and addresses benefits and challenges of these algorithms which would be beneficial for beginners..

No abstract

Grass evidence often encountered at the crime scene, helps in establishing the primary as well as secondary crime scenes. Due to the limited quantity and intricate nature of samples, there is a need for fast, sensitive and reliable techniques for the effective analysis of this vital evidence. In the current study, an attempt is made to study the feasibility of ATR FT-IR in combination with chemometric tools for chemical characterization and species differentiation of grass stains. Ten different grass species belonging to three subfamilies, Panicoideae, Chlorodoidae, and Bambusoideae, were collected and analyzed by using the ATR FT-IR spectroscopy combined with the PLSR chemometric tool. Results showed a clear difference between samples of Panicoideae, Chlorodoidae, and Bambusoideae subfamilies, and between species of each subfamily. To analyse the performance of the classification model, a set of 10 unknown/blind samples (unknown to the analyst) were randomly selected from the training dataset and all unknown/blind samples were accurately assigned in their corresponding group. ATR FT-IR spectroscopy requires minimal sample preparation and can be successfully used as an eco-friendly, non-destructive, and reliable alternative to other existing methods.

No abstract