China
Africa
Explore chapters and articles related to this topic
Big Data in Medical Image Processing
Published in R. Suganya, S. Rajaram, A. Sheik Abdullah, Big Data in Medical Image Processing, 2018
R. Suganya, S. Rajaram, A. Sheik Abdullah
In this segment, the different methods that are frequently used for data classification will be discussed. The most common methods used in data classification are decision trees, Support Vector Machine methods, Naive Bayesian method, instance-based method and neural networks. The classification among technologies is illustrated in Figure 2.
Vehicular Social Networks
Published in Yufeng Wang, Athanasios V. Vasilakos, Qun Jin, Hongbo Zhu, Device-to-Device based Proximity Service, 2017
Yufeng Wang, Athanasios V. Vasilakos, Qun Jin, Hongbo Zhu
Machine learning techniques are extensively used for such data classification. This platform collects standard vehicle information such as the speed, pressure on the brake or gas pedal, steering wheel rotation, and global positioning system (GPS) routing.
Efficient Random Forest Algorithm for Multi-objective Optimization in Software Defect Prediction
Published in IETE Journal of Research, 2023
Shailza Kanwar, Lalit Kumar Awasthi, Vivek Shrivastava
The proposed method aims to analyze the class imbalance problems for SDP. In recent days, several real-time applications are revolutionizing today’s environment by coordinating with technological developments. Data classification becomes a challenging task when it is bounded by huge data sizes and the imbalanced nature of data. Class imbalance issues play an important role in data-mining tasks. The data samples that belong to more classes than other classes are known as class imbalance issues. The survey reports that the data-mining algorithms focus on the major classes rather than the minority classes. Although minority classes occur rarely, they are significant. Here, the feature selection approach is more capable of improving class imbalance issues than the data pre-processing approach.
A data classification method based on particle swarm optimisation and kernel function extreme learning machine
Published in Enterprise Information Systems, 2023
Ao Liu, Dongning Zhao, Tingjun Li
According to the characteristics of the known cloud data set, data classification is a technique to construct a classifier, and to assign classes to the samples of unknown categories by using the classifier(Roy et al. 2017; Yen et al. 2015). Training and testing are included in the classifier, which can classify sample data effectively (Sun et al. 2016). The characteristics of data set are summarised and analysed during the training phase, and in the testing stage, the classification accuracy is tested by using category description or model. There are many different algorithms and models for data classification, such as support vector machine (SVM), deep learning, genetic algorithm (GA), ELM and so on. With the above-mentioned models, classification algorithms have been applied in business activities, financial management, the Internet, electronic information, and so on.
An ensemble machine learning method for crash responsibility assignment in quasi-induced exposure theory
Published in Journal of Transportation Safety & Security, 2023
Guopeng Zhang, Ying Cai, Xinguo Jiang, Yingfei Fan, Yue Zhou, Jun Qian
The crash responsibility assignment is essentially a classification problem. That is, the crash-involved drivers need to be correctly classified into two categories, including responsible and non-responsible parties. In this regard, machine learning is a powerful technique that is commonly employed for data classification. Through data training, it can predict the category that each driver belongs to, with the given relevant factors. Compared to the conventional statistical regression, machine learning models are more flexible in the modeling exercise and can achieve a higher accuracy, so that it has gradually witnessed popularity in the field of traffic safety research in recent years. Thus, the study attempts to employ machine learning methods to achieve the crash responsibility assignment.