Khalifa University

About: Khalifa University is a education organization based out in Abu Dhabi, United Arab Emirates. It is known for research contribution in the topics: Computer science & Adsorption. The organization has 3752 authors who have published 10909 publications receiving 141629 citations.

Developing an online hate classifier for multiple social media platforms

Abstract: The proliferation of social media enables people to express their opinions widely online. However, at the same time, this has resulted in the emergence of conflict and hate, making online environments uninviting for users. Although researchers have found that hate is a problem across multiple platforms, there is a lack of models for online hate detection using multi-platform data. To address this research gap, we collect a total of 197,566 comments from four platforms: YouTube, Reddit, Wikipedia, and Twitter, with 80% of the comments labeled as non-hateful and the remaining 20% labeled as hateful. We then experiment with several classification algorithms (Logistic Regression, Naive Bayes, Support Vector Machines, XGBoost, and Neural Networks) and feature representations (Bag-of-Words, TF-IDF, Word2Vec, BERT, and their combination). While all the models significantly outperform the keyword-based baseline classifier, XGBoost using all features performs the best (F1 = 0.92). Feature importance analysis indicates that BERT features are the most impactful for the predictions. Findings support the generalizability of the best model, as the platform-specific results from Twitter and Wikipedia are comparable to their respective source papers. We make our code publicly available for application in real software systems as well as for further development by online hate researchers.

From cotton to wearable pressure sensor

Abstract: In this work, carbon cottons (CC) with moderate electrical conductive (11 S m−1) were prepared from cotton via a simple pyrolysis process. Flexible and electrical conductive CC/polydimethylsiloxane (PDMS) composites were fabricated by vacuum assisted infusion of PDMS resin into a CC scaffold. Based on the CC/PDMS composites prepared, a simple yet highly sensitive pressure sensor was developed, which shows a maximum sensitivity of 6.04 kPa−1, a wide working pressure up to 700 kPa, a wide response frequency from 0.01 to 5 Hz, and durability over 1000 cycles. Based on our knowledge, the pressure sensitivity of the CC/PDMS sensor is only next to the record value in a pressure sensor (8.4 kPa−1). By integrating the pressure sensor with a sport shoe and waist belt, we demonstrate that the real time sport performance and health condition could be monitored. Notably, the device fabrication process is simple and scalable with low-cost cotton as raw material. The CC/PDMS composites are believed to have promising potential applications in wearable electronic devices such as, human-machine interfacing devices, prosthetic skins, sport performance, and health monitoring.

Secure and Robust Machine Learning for Healthcare: A Survey

Abstract: Recent years have witnessed widespread adoption of machine learning (ML)/deep learning (DL) techniques due to their superior performance for a variety of healthcare applications ranging from the prediction of cardiac arrest from one-dimensional heart signals to computer-aided diagnosis (CADx) using multi-dimensional medical images. Notwithstanding the impressive performance of ML/DL, there are still lingering doubts regarding the robustness of ML/DL in healthcare settings (which is traditionally considered quite challenging due to the myriad security and privacy issues involved), especially in light of recent results that have shown that ML/DL are vulnerable to adversarial attacks. In this paper, we present an overview of various application areas in healthcare that leverage such techniques from security and privacy point of view and present associated challenges. In addition, we present potential methods to ensure secure and privacy-preserving ML for healthcare applications. Finally, we provide insight into the current research challenges and promising directions for future research.