China
Africa
Explore chapters and articles related to this topic
Robotics and Machine Learning
Published in Shivani Agarwal, Sandhya Makkar, Duc-Tan Tran, Privacy Vulnerabilities and Data Security Challenges in the IoT, 2020
This permits us to precisely and accurately generate output when given new inputs. There are two types of supervised learning: classification and regression. Classification is a type of supervised learning which predicts the results of given input sample when the output is in the shape of classification. A classification model might see the input data and attempt to anticipate names like “sick” or “healthy.”Regression is another type of supervised learning which predicts the results of given input sample when the output is in the shape of real values. For illustration, a relapse exhibit might handle input data to anticipate the sum of precipitation like rainfall or a person’s height and so on.
Machine Learning Classifiers
Published in Rashmi Agrawal, Marcin Paprzycki, Neha Gupta, Big Data, IoT, and Machine Learning, 2020
Usually, the term “machine learning” is interchangeable with artificial intelligence, however machine learning is in fact an artificial intelligence sub-area. It is also defined as predictive analysis or predictive modeling. Defined in 1959 by Arthur Samuel, an American computer scientist, the term “machine learning” is the ability of a computer to learn without explicit programming. To predict output values within a satisfactory range, machine learning uses designed algorithms to obtain and interpret input data. They learn and optimise their operations as new data is fed into these algorithms to enhance performance and develop intelligence over time. At present, there are several categories of algorithms for machine learning and they are largely classified as supervised, semi-supervised, unsupervised and reinforcement. Classification is the supervised learning process where classes are sometimes referred to as targets/labels or categories to predict the class of given data points. The machine learning programs draw conclusions in classification from given values and find the category to which new data points pertain. For example, in the context of spam and non-spam classification of emails, the program works on existing data (emails) and filters out the emails as “spam” or “not spam”.
Regression
Published in Dirk P. Kroese, Zdravko I. Botev, Thomas Taimre, Radislav Vaisman, Data Science and Machine Learning, 2019
Dirk P. Kroese, Zdravko I. Botev, Thomas Taimre, Radislav Vaisman
Francis Galton observed in an article in 1889 that the heights of adult offspring are, on the whole, more “average” than the heights of their parents. Galton interpreted this as a degenerative phenomenon, using the term “regression” to indicate this “return to mediocrity”. Nowadays, regression refers to a broad class of supervised learning techniques where the aim is to predict a quantitative response (output) variable y via a function g(x) of an explanatory (input) vector x = [x1, …, xp]⊤, consisting of p features, each of which can be continuous or discrete. For instance, regression could be used to predict the birth weight of a baby (the response variable) from the weight of the mother, her socio-economic status, and her smoking habits (the explanatory variables). regression
Evaluation of seawater intake discharge coefficient using laboratory experiments and machine learning techniques
Published in Ships and Offshore Structures, 2023
Mahmood Rahmani Firozjaei, Seyed Taghi Omid Naeeni, Hassan Akbari
Data-mining techniques and statistical analyses can be used to discover the relationships between different variables. Various techniques have been applied to solve these engineering problems. Two models, regression analysis and model tree (MT), were used to evaluate each model’s accuracy in predicting the discharge coefficient (Cd). These methods are briefly discussed below. More details can be found in the related references. The two major parts of supervised learning methods are classification and regression problems. There are different types of Regression analysis, and linear and nonlinear analyses are commonly used for regression. Nonlinear regression is sometimes more applicable than linear regression and can describe the relationship between variables properly.
A review on the applications of machine learning and deep learning in agriculture section for the production of crop biomass raw materials
Published in Energy Sources, Part A: Recovery, Utilization, and Environmental Effects, 2023
Wei Peng, Omid Karimi Sadaghiani
In supervised learning, data scientists describe the variables they want the algorithm to look for connections between and provide the algorithms with labeled training data. The algorithm’s input and output are both described. Unsupervised learning: Algorithms trained on unlabeled data are used in this sort of machine learning. The program searches through data sets in search of any significant relationships. Both the input data that algorithms use to train and the predictions or suggestions they provide are predefined. Semi-supervised learning is a method of machine learning that combines the two categories mentioned above. An algorithm may be fed mostly labeled training data by data scientists, but the algorithm is allowed to explore the data on its own and come to its own conclusions about the data set. Data scientists often use reinforcement learning to instruct a computer to carry out a multi-step procedure for which there are set rules. An algorithm is programmed by data scientists to fulfill a goal, and they provide it with positive or negative feedback as it determines how to do so. However, the algorithm often chooses the course of action on its own.
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.