China
Africa
Explore chapters and articles related to this topic
Artificial Intelligence and Machine Learning
Published in Shahab D. Mohaghegh, Smart Proxy Modeling, 2023
Currently, Artificial Intelligence is activated through Machine Learning. Machine Learning is the science of making computers to act without being explicitly programed. Machine Learning is done through Open Computer Algorithms and learning from data. “It a set of algorithms that parse data, learn from data, discover patterns in the data, to make predictions or decisions based on what they have learned” [Khan Academy, https://www.khanacademy.org/]. “There are situations in which the learning process takes place by establishing a relationship between input and output labeled data. This learning technique is known as supervised learning. On the other hand, unsupervised learning is a more intuitive approach in which the algorithm can find structure or patterns in unlabeled data through clustering methods, among others” [Khan Academy].
Use of Machine Learning in Healthcare
Published in Punit Gupta, Dinesh Kumar Saini, Rohit Verma, Healthcare Solutions Using Machine Learning and Informatics, 2023
Machine learning algorithms are broadly split into supervised and unsupervised learning. Both categories comprise a variety of algorithms that are used to implement mathematical models.Supervised learning comprises labeled training data. They mainly focus on classification and regression problems. Some examples of supervised learning algorithms are: random forest, decision trees, naïve Bayes models, and support vector machine (SVM).Unsupervised learning uses unlabeled data for model training. The most common algorithms for unsupervised learning are K-means clustering and deep learning.
Application of Image Processing and Data in Remote Sensing
Published in Ankur Dumka, Alaknanda Ashok, Parag Verma, Poonam Verma, Advanced Digital Image Processing and Its Applications in Big Data, 2020
Ankur Dumka, Alaknanda Ashok, Parag Verma, Poonam Verma
Thus, clustering is an approach to segregate groups with similar traits and assign them into clusters. Thus, we can say that clustering is one of the popular techniques used in unsupervised learning where grouping of data is done based on the similarity of the data points. Clusters can be formed using the concept that objects with similarities can be grouped together. Similarly, clustering which is an unsupervised learning method can be considered as a process that helps the machines to distinguish different objects given in a dataset. Since clustering is an unsupervised learning, no external labels are attached to the given instances and machines find clusters based on the patterns or the trends observed. Various algorithms are used to extract the parameters that can help to group instances into appropriate clusters. Clustering learning mode helps to divide the data into different groups as classes such that each data point is similar to data points in the same group and dissimilar to the data points in the other groups. Clustering is considered to be a useful technique to differentiate groups from the unlabeled datasets.
Intelligent structural health monitoring of composite structures using machine learning, deep learning, and transfer learning: a review
Published in Advanced Composite Materials, 2023
Muhammad Muzammil Azad, Sungjun Kim, Yu Bin Cheon, Heung Soo Kim
Machine learning (ML) involves the development of algorithms that establish relations between variables based on the existing data, and can be utilized for classification, regression, or density estimation [62]. The goal of classification is to define the decision boundary that can discriminate the input data into distinct health states. Regression is associated with developing a computational model to predict continuous numerical values, such as remaining useful life (RUL). In comparison, density estimation involves the approximation of the underlying probability density function (PDF) from a set of data points with no target values or output. Generally, classification and regression problems are carried out using a supervised learning approach, whereas density estimation is dealt with using an unsupervised approach. Supervised learning algorithm involves labelled data to make predictions on the future unseen data, while unsupervised learning aims to identify underlying patterns or structures in data, without requiring pre-defined labels for the observations. Figure 4 illustrates the primary categories of ML and some of its prevalent algorithms.
A comparative study between PCR, PLSR, and LW-PLS on the predictive performance at different data splitting ratios
Published in Chemical Engineering Communications, 2022
For unsupervised machine learning, only the input will be provided during training and the objective of the model is to determine underlying structures or distributions within the data (McCue 2015). Unsupervised learning can be further grouped into clustering and dimensionality reduction (Vieira et al. 2020). Examples of unsupervised learning techniques include partitional clustering, hierarchical clustering and principal component analysis (PCA) (Talabis et al. 2015; Sato et al. 2019). In more recent years, these types of models have also been prevalent in deep learning techniques such as stacked auto-encoders as seen in the research work of Yuan et al. (2020) and Wu et al. (2021). Moreover, some review studies on the application of these machine learning in prognosis and fault diagnosis as well as process monitoring, control and optimization can be found in Jiang and Yin (2019) and Jiang et al. (2021).
A distributed unsupervised learning algorithm and its suitability to physical based observation
Published in International Journal of Parallel, Emergent and Distributed Systems, 2022
Unsupervised learning techniques are the set of machine learning algorithms useful when little or no labelled data is provided. Conventional techniques including K-Means [1] and variant algorithms [2,3] are confined to a predetermined number of classes present in the dataset which is unsuitable for investigatory or ‘self-driven’ learning. Techniques such as Density Based Spatial Clustering (DBScan) [4] relax this constraint but fail to resolve clusters where class boundaries are typically ill-defined. These may be caused from merging of multiple classes, no matter how sparse the overlapping tails, with outliers potentially leading to clustering where no conglomeration should exist. In either existing techniques, with large data sets, clustering becomes un-computable or not readily feasible because of the tight coupling of the information required by the algorithms. Computational complexity, storage, memory requirements and distributability of algorithms become bottlenecks limiting the usefulness of any learning technique in large data applications.