A Non-Verbal Facial Sentiment Analysis Using Convolutional Neural Network.

• Building a computer vision and natural language based application to analysis the basic seven non-verbal facial sentiment with wide area of application including physiology, psychology, neuroscience, security monitoring, and augmented reality etc.

EnPhis: An ensemble learning based phishing website detection algorithms.

• Phishing is an illegal act by which the phisher attempts to entice people to provide essential private information like user IDs, bank account details, passwords. An increasing number of such attack incidents are exposed globally at the personal and organizational levels and are typically triggered by emails, instant messages, or phone calls. There are lot of studies where researchers attempted to find out an effective approach for detecting phishing websites. However, no particular approach is available which has efficiently detected phishing websites due to higher rate of false positive and false negative during the time of detecting legitimate and non-legitimate websites. Instead of using traditional machine learning(ML) approaches, in this paper, we propose a new approach based on ML technique which uses multiple classifiers. Our propose model able to predict efficiently which one is phishing and which one is not. Then, we compare the performance of propose model with six traditional ML-based classifiers and three deep learning(DL) based classifiers. Additionally, this article also try to show the comparison of the performance metrics between before and after feature selection. And the result shows that our proposed approach provide better performance metrics to predict phishing and benign websites on both cases (before and after feature feature selection) than other traditional ML and DL models.