Amel Ben Mahjoub

Papers

Total Citations

H-Index

About

Amel Ben Mahjoub is a researcher specializing in deep learning and human activity recognition, with a focus on developing efficient, end-to-end architectures for sensor-based classification. Her most-cited work, "An efficient end-to-end deep learning architecture for activity classification" (2018, 21 citations), introduces a streamlined neural network design that bypasses traditional feature engineering, directly learning from raw data to classify complex human activities. This contribution addresses a critical bottleneck in ubiquitous computing—balancing model accuracy with computational efficiency for real-time applications. By demonstrating that a compact, fully convolutional architecture can outperform more complex, multi-stage methods, Ben Mahjoub’s research has influenced subsequent work in wearable and mobile sensing, where resource constraints are paramount. Her approach not only simplifies deployment but also enhances generalizability across diverse activity datasets. As a researcher at the intersection of machine learning and human-computer interaction, Ben Mahjoub continues to push toward more adaptive, lightweight models that can operate seamlessly in everyday environments, making her work a valuable reference for students and engineers aiming to bridge the gap between algorithmic innovation and practical, on-device intelligence.