k. dhanashree

Each and every human have unique fingernails. In the early days, the psychological conditions of the human body were reflected with the help of the growth situation of the surface of nails.  It is possible to diagnose human nails and predict the disease. Predicting the disease at the early stage helps in preventing the disease. In this proposed work, the image of the nail is taken from a microscopic image. The lunula and nail plate are segmented effectively using the image pre-processing techniques. Histograms of oriented gradients and local binary patterns are used to capture the characteristic value. Once after pre-processing various features of the nails are extracted using various machine learning algorithms such as Support Vector Machines, Multiclass Support Vector Machine, Convolution Neural Network along with an Optimization algorithm named Ant Colony Optimization to improve the efficiency of classification.

Digital currency plays a vital role in the process of trading as it helps the sellers and buyers to earn more profit. In today’s world, many categories of cryptocurrencies exist and each one of them employs its own security algorithms. Bitcoin price prediction is a complex problem that needs advanced algorithms to solve exactly. In this paper, swarm-based intelligence algorithms are applied in order to solve the bitcoin value prediction problem. In particular, Ant Colony Optimization and Binary Grasshopper Optimization algorithms are combined as a hybrid framework to select the most critical features in the dataset for bitcoin value prediction. The extracted features from the hybrid model are given as input to the convolutional neural network to predict the price of the bitcoins. As per the experimental results, the proposed hybrid algorithm produces better results when compared with the stand-alone version of grasshopper and neural network algorithms.