Machine learning
Researchers from Tezpur University in Assam, India, have developed a novel technique to detect epileptic seizures using machine learning and deep learning algorithms on electroencephalography (EEG) inputs. The research, published in the journal Bioengineering, aims to improve the efficiency and accuracy of epilepsy diagnosis and treatment. The team, led by
Machine learning
Researchers at Shandong University have developed a novel framework, known as Federated Weighted Learning Algorithm (FWLA), to mitigate poisoning attacks in distributed machine learning. According to the study, existing defense methods are often rigid and lack adaptability to dynamic adversarial behaviors, leading to high computational costs. FWLA, on the other
Machine learning
Research published by investigators from Baskent University evaluates the performance of large language models (LLMs) on questions from the Turkish Medical Specialty Training Entrance Exam. The study assessed four LLMs: OpenAI-ChatGPT 4, Google AI-Gemini 1.5 Pro, Cohere-Command R +, and Meta AI-Llama 3 70B. The results showed that ChatGPT 4
Machine learning
The global artificial intelligence in drug discovery market has witnessed significant growth in recent years as pharmaceutical companies increasingly adopt AI to accelerate and optimize drug discovery processes. According to a report by ResearchAndMarkets.com, the market was valued at USD 1.72 billion in 2024 and is projected to
Machine learning
In a move to enhance business decision-making and growth, Jesta I.S. has launched Vision Ask Jane, a pioneering AI-powered growth advisor designed for retailers, wholesalers, and brands. This innovative tool combines artificial intelligence (AI), machine learning (ML), and natural language processing (NLP) to provide tailored insights and recommendations for
Machine learning
Researchers at the International Hellenic University have developed an innovative educational approach to introduce students to core AI and ML concepts through a robotics-based lesson. This approach has been found to increase student engagement and enthusiasm, as well as improve conceptual understanding. Key Takeaways: * The study presents a design and
Machine learning
In a significant breakthrough, a wearable computing apparatus has been designed to overcome limitations in current augmented, virtual, and artificial intelligence (AI) technologies. The innovative device, invented by Olaoluwa O. Adesanya, enables users to interact with objects in immersive environments, leveraging advancements in computing, sensors, and haptic feedback. This wearable
Machine learning
Wearable sensors and artificial intelligence have been successfully integrated to analyze physical activities in the sport of alpine skiing. The research, led by the University of Chemistry and Technology in Prague, employed digital signal processing, numerical methods, and machine learning to evaluate ski movement patterns. The findings indicate that the
Machine learning
Broadridge Financial Solutions, a leading fintech and technology company, has been awarded a new U.S. patent for its large language model (LLM) orchestration of machine learning agents. This patented method is behind BondGPT, an award-winning GenAI application that provides timely, secure, and accurate responses to natural language questions. The
Machine learning
Researchers from Iqra University in Karachi, Pakistan, have made a significant breakthrough in software quality assurance using artificial intelligence. The study focuses on developing a novel transformer-based model, Relation-Aware BERT (RABERT), to enhance automated code smell detection. The research aims to improve the accuracy and efficiency of code analysis, enabling
Machine learning
Air quality has become a significant environmental concern globally due to the rapid development of society and industry. Predicting and rating air quality for many cities remains a challenge. Recent research from the University of Hull has explored the potential of machine learning algorithms in addressing this issue. By developing
Machine learning
Researchers at Guangdong University of Science and Technology have proposed innovative semi-supervised attribute selection algorithms for partially labeled multiset-valued data, aiming to overcome the limitations of traditional machine learning approaches. The study highlights the importance of addressing incomplete information within a system, which can be effectively tackled through the application
