China
Africa
Explore chapters and articles related to this topic
Evaluation of factors affecting long-term creep of concrete using machine learning regression models
Published in Joan-Ramon Casas, Dan M. Frangopol, Jose Turmo, Bridge Safety, Maintenance, Management, Life-Cycle, Resilience and Sustainability, 2022
Creep is a complex phenomenon that is affected by several factors such as concrete mixtures (i.e., water-cement ratio, compressive strength, type of cement, and admixture) (Chunping et al. 2019, Daou & Rapahel 2021, Daou & Raphael 2020, Kamen et al. 2009, Shen et al. 2017), environmental conditions (i.e., temperature, and relative humidity) (Briffaut et al. 2012, Theiner et al. 2017, Zheng et al. 2020), and loading schemes (i.e., age at loading, and load) (Shen et al. 2020, Su et al. 2017). These tests provided valuable data for understanding the creep behavior of concrete. However, an accurate prediction remains a challenge because of the high sensitivity of creep to wide ranges of parameters and the interplay between these parameters. This study provides a Machine learning (ML) solution to predict creep coefficient at long-term and evaluate the factors affecting it. As a tributary of artificial intelligence, supervised ML has been considered as a promising data-driven approach to build robust models for predicting various properties and behaviors of heterogeneous materials.
Investigation of IoMT-Based Cancer Detection and Prediction
Published in Meenu Gupta, Rachna Jain, Arun Solanki, Fadi Al-Turjman, Cancer Prediction for Industrial IoT 4.0: A Machine Learning Perspective, 2021
Meet Shah, Harsh Patel, Jai Prakash Verma, Rachna Jain
ML is a subset of artificial intelligence (AI) and consists of algorithms that learn automatically through experience. DL is a part of ML based on an artificial neural network (ANN) that mimics the function of the human brain. Both ML and DL techniques play an important role in cancer detection and prediction [16–20]. These ML/DL techniques usually work by segmenting or detecting abnormalities and classifying the segmented area or tumor into benign or malignant. On the other hand, these techniques might also be trying to predict the risk of cancer to a patient based on his/her diagnostic history rather than detecting if he/she has cancer or not. DL methods can automatically learn features from large amounts of raw images and perform cancer detection and classification. The use of ML/DL techniques for cancer detection allows us to move from a reactive approach to a more proactive, personalized approach for cancer detection and prediction.
Malware Detection and Mitigation
Published in Nicholas Kolokotronis, Stavros Shiaeles, Cyber-Security Threats, Actors, and Dynamic Mitigation, 2021
Gueltoum Bendiab, Stavros Shiaeles, Nick Savage
With the rapid growth and evolution of malicious code, analysis and detection of malware based on static and dynamic analysis tools become insufficient and have compelled researchers to derive novel analysis and detection solutions. Machine learning (ML) is among the innovative and successful technologies that have been employed toward that direction. ML is a branch of artificial intelligence (AI) that uses a collection of methods and algorithms, which emulate human intelligence by learning from the surrounding environment. It was defined by Arthur Samuel as “a field of study that gives computers the ability to learn without being explicitly programmed” [31]. More specifically, ML algorithms have the ability to identify specific trends and patterns from large volumes of data without prior knowledge or human interventions. In fact, these algorithms have demonstrated great success in learning complex patterns that enable them to make accurate predictions about unobserved data [32].
Current applications and future impact of machine learning in emerging contaminants: A review
Published in Critical Reviews in Environmental Science and Technology, 2023
Lang Lei, Ruirui Pang, Zhibang Han, Dong Wu, Bing Xie, Yinglong Su
ML is a subset of artificial intelligence (AI) that emphasizes mimicking human learning by using data and algorithms and then gradually improving its accuracy. Over the last two decades, ML has progressed from a laboratory curiosity to a viable commercial technology (Jordan & Mitchell, 2015). Generally, ML is considered to contain three main categories: classical ML, deep learning, and reinforcement learning, and plentiful algorithms are developed (Figure 2). ML evolved from AI since 1980, and deep learning is an emerging category that realizes the original goal of AI. In the past decade, reinforcement learning has also emerged as a powerful approach within ML, enabling agents to learn optimal policies by interacting with their environment and receiving feedback in the form of rewards or penalties.
Ergonomic investigations on novel dynamic postural estimator using blaze pose and transfer learning
Published in Ergonomics, 2023
Vigneswaran Chidambaram, Madhan Mohan Gopalsamy, Vignesh Raja M, Brajesh Kumar Kanchan
ML is an Artificial Intelligence (AI) subset that aids software in improving prediction accuracy. Furthermore, the TL method is a subset of ML. ML algorithms use historical data to anticipate new output values. Machine Learning (ML) technology would be a suitable substitute for building this proposed work (Chan et al. 2022). Various human pose estimation algorithms are available in ML (Mroz et al. 2021). The acquired human postural data would decide the RULA action levels. The camera connected to a computer can capture the human body posture landmarks with RGB colour values. The human body can be represented as a kinematic model, a planar model, and the volumetric model. The modelled human for postural assessment could be in either 2D or 3D. However, the planar model has a restriction compared to a 2D model. In the case of volumetrics, it is limited to the 3D model. Still, both the 2D and 3D models can be built using the kinematic model.
The quest for business value drivers: applying machine learning to performance management
Published in Production Planning & Control, 2022
Franco Visani, Anna Raffoni, Emanuele Costa
Research in this field is relatively recent and whilst growing at a fast pace, is still theoretically and practically underdeveloped (Möller, Schäffer, and Verbeeten 2020; Appelbaum et al. 2017; Schneider et al. 2015). In particular, the application of computer-based algorithms like Machine Learning (ML) is relatively unexplored (Nielsen 2022), despite their increased pervasiveness in organisational life and implications for managerial decision making (Moll and Yigitbasioglu 2019). ML is an application of artificial intelligence (AI) that provides systems with the ability to automatically learn and improve from experience without being explicitly programmed. The process of learning begins with observations or data, to search for patterns in past data and support more effective future decisions (Mitchell 1997). ML and statistics are similar in many aspects but while statistical analysis is grounded in probability theory and distributions, ML is a set of mathematical functions, iteratively optimised, that are combined to best predict an outcome (Witten and Frank 2002). The algorithm can be ‘supervised’ when applied to already-labelled data or ‘unsupervised’ when the information used to train is neither classified nor labelled (Alloghani et al. 2020).