Muhammad Osama

BTech(MechEng)

Mechanical Engineering with specialization in AI & Robotics

Muhammad Osama is a full-time data analytics consultant and freelance technical writer based in Delhi, India. He specializes in transforming complex technical concepts into accessible content. He has a Bachelor of Technology in Mechanical Engineering with specialization in AI & Robotics from Galgotias University, India, and he has extensive experience in technical content writing, data science and analytics, and artificial intelligence. Muhammad has worked on various research projects related to data analytics, machine learning, and deep learning for industries such as retail, medical, finance, agriculture, and Ed-Tech. He is enthusiastic and passionate about Artificial Intelligence-related research. Outside of work, he enjoys learning about new developments across all the fields related to science, technology, engineering, mathematics, and statistics.

Articles from Muhammad

Pedipulate: Quadruped Robot Manipulation Using Legs

Swiss researchers introduce Pedipulate, a novel controller trained with deep reinforcement learning, enabling quadruped robots to manipulate objects using their legs. Demonstrating robustness and versatility, Pedipulate tracks foot target points, adapts stance, and handles external disturbances, showcasing potential applications in maintenance, home support, and exploration tasks.

25 Feb 2024

Preference Alignment Framework for Enhancing Multi-Modal Large Language Models

This research introduces a novel preference alignment framework to address performance degradation in multi-modal large language models (MLLMs) caused by visual instruction tuning. By leveraging preference data collected from a visual question answering dataset, the proposed method significantly improves the MLLM's instruction-following capabilities, surpassing the performance of the original language model on various benchmarks.

23 Feb 2024

Harnessing Machine Learning for Predicting School Dropout

Researchers utilized various machine learning algorithms to develop predictive models for identifying students at risk of dropping out of secondary and higher education in Mexico. Leveraging demographic, socioeconomic, and educational data, the study demonstrated the effectiveness of artificial neural networks (ANN) in achieving high reliability (99%) in predicting school dropout, highlighting key variables such as school attendance, type, location, occupation, income, and marital status.

22 Feb 2024

Optimizing Content Caching in Vehicular Networks: A Gale–Shapley Approach

Researchers proposed a novel content caching scheme for fog nodes in 5G-enabled vehicular networks, utilizing a modified Gale–Shapley algorithm to optimize content placement based on the preferences of content providers and vehicles. Their approach effectively reduced content downloading time, improved quality of service, and ensured fair and efficient content placement by considering factors such as content popularity, vehicle connectivity, and channel quality.

21 Feb 2024

Urban Park Management Through Social Media and Deep Learning

Researchers propose a groundbreaking framework utilizing social media data and deep learning techniques to assess urban park management effectively. By analyzing visitor comments on seven parks in Wuhan City, the study evaluated various management aspects and identified improvement suggestions, demonstrating the potential of this approach to enhance park service quality and management efficiency. The framework's dynamic visualization capabilities and scalability make it a valuable tool for improving public spaces and contributing to the development of smart cities, with opportunities for expansion to other urban areas and data sources in future research.

20 Feb 2024

Optimizing Electric Vehicle Power: A Novel Energy Management Strategy

Researchers present an innovative energy management strategy (EMS) for electric vehicles (EVs), combining slap swarm optimization and differential flatness control to enhance performance and reliability. The proposed EMS efficiently coordinates power flow among battery, supercapacitor, and motor components, stabilizing voltage, reducing harmonics, and meeting power demand, promising improved EV efficiency and driving range while reducing greenhouse gas emissions.

18 Feb 2024

AI's Deep Impact: Exploring Facial Expressions in Job Interviews

Researchers harnessed AI technology to create deepfake videos portraying various facial expressions, investigating their influence on observer perceptions in job interviews. The study highlights how deepfake facilitates controlled experimentation in studying nonverbal behavior, shedding light on its crucial role in social interactions and offering insights for job interview training and beyond.

18 Feb 2024

Analytical Estimation of Robotic Compact Storage and Retrieval System Performance

Researchers from Austria introduce an analytical method to estimate the performance of Robotic Compact Storage and Retrieval Systems (RCS/RS), crucial for modern storage operations. Their approach, based on cycle time models and validated through simulation, offers insights into system configurations and operation modes, aiding in optimal design and operation decisions. The study establishes a benchmark for comparing RCS/RS performance and provides a valuable tool for material handling and logistics practitioners.

16 Feb 2024

Real-time Water Quality Monitoring and Prediction System using IoT and Cloud Computing

Researchers developed a comprehensive system leveraging IoT and cloud computing to monitor and predict drinking water quality in real-time. The system integrates sensors, microcontrollers, web servers, and machine learning models to collect, transmit, analyze, and predict water quality parameters. Machine learning algorithms, particularly decision trees, achieved high accuracy in predicting drinkability, demonstrating the system's potential to enhance water safety and contribute to achieving Sustainable Development Goals.

16 Feb 2024

FlashNet: Advancing Lightning Forecasting with Hybrid AI

Researchers developed FlashNet, a hybrid AI method, to forecast lightning flashes up to 48 hours ahead, surpassing traditional NWP models. Utilizing features from high-resolution NWP data and employing deep neural networks, FlashNet demonstrated superior accuracy, reliability, and sharpness, offering valuable insights for various sectors vulnerable to lightning-related risks. The study highlights FlashNet's potential for medium-range forecasting and recommends further exploration for extending forecast horizons and addressing global applicability.

16 Feb 2024