In the last decade, Deep neural networks (DNNs) have been proven to outperform conventional machine learning models in supervised tasks. Most of these are typically optimized by minimizing well-known Cross-Entropy objective function. The latter, however, has a number drawbacks, including poor margins and instability. Taking inspiration from recent self-supervised Contrastive representation appr...

In this paper, we introduce a framework allowing users to conceptualize by both constructing and supervising the evolution of an ontology. Because of the various discussions around the terms ontology and conceptualization in the knowledge sharing community we will firstly show, by means of a simple example, how these terms are apprehended by our framework. We will then discuss philosophical asp...

Most learning algorithms for factor graphs require complete inference over the dataset or an instance before making an update to the parameters. SampleRank is a rank-based learning framework that alleviates this problem by updating the parameters during inference. Most semi-supervised learning algorithms also rely on the complete inference, i.e. calculating expectations or MAP configurations. W...

We present a framework to address the imbalanced data problem using semi-supervised learning. Specifically, from a supervised problem, we create a semi-supervised problem and then use a semi-supervised learning method to identify the most relevant instances to establish a welldefined training set. We present extensive experimental results, which demonstrate that the proposed framework significa...

Semi-supervised learning is crucial in many applications where accessing class labels unaffordable or costly. The most promising approaches are graph-based but they transductive and do not provide a generalized model working on inductive scenarios. To address this problem, we propose generic framework, ESA☆, for semi-supervised based three components: an ensemble of autoencoders providing new d...

Today, the term ransomware is frequently used in cybercrime headlines, its consequences have been on rise leaving a trail of terrible losses wake. Both people and businesses victimized by ransomware, costing victims millions dollars ransom payments. In addition, who were unable to pay or decrypt data experienced losses. This study uses dynamic malware analysis artifacts supervised machine learn...

Atlantic is an open-source Python package designed to simplify and automate data preprocessing for supervised ML (Machine Learning) tasks. The integrates multiple customizable mechanisms, including datetime feature engineering, automated selection, categorical encoding techniques null imputation methods. In order provide a comprehensive approach processing automation, pipeline follows optimizat...