China
Africa
Explore chapters and articles related to this topic
Thematic Map Accuracy Assessment Considerations
Published in Russell G. Congalton, Kass Green, Assessing the Accuracy of Remotely Sensed Data, 2019
Russell G. Congalton, Kass Green
If possible, it is also advantageous to use a classification scheme that is hierarchical. A hierarchical classification scheme typically has many levels, beginning with the most general categories at Level 1 and growing in detail when moving to higher levels. For example, Level 1 classes might include forest, developed, water, and so on, while the Level 2 forest classes might be divided into coniferous and hardwood (Figure 6.2). In hierarchical systems, specific categories within the classification scheme can be collapsed to form more general categories. This ability is especially important when it is discovered that certain map categories cannot be reliably mapped. For example, it may be impossible to separate interior live oak from canyon live oak in California’s oak woodlands (these two oak types are almost indistinguishable on the ground). Therefore, these two categories may have to be collapsed to form a live oak class that can be reliably mapped.
Image Analysis and Retrieval via Self-Organization
Published in Kim-Hui Yap, Ling Guan, Stuart William Perry, Hau-San Wung, Adaptive Image Processing, 2018
Kim-Hui Yap, Ling Guan, Stuart William Perry, Hau-San Wung
Within a given period of the hierarchy control function, the learning tends to be a stochastic process, the weight vectors will eventually converge in the mean-square sense to the probabilistic mean of their corresponding input vectors as the learning rate α(t) decreases. The hierarchical classification organizes the samples such that each node represents a subset of samples which share some similar features. Such hierarchical classification can be quickly searched to find a matching pattern for a new input.
Classification
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
In hierarchical classification a hierarchical relation between classes/labels is taken into account during the classification process. Usually, the relations are modeled via a tree or a directed acyclic graph. A visual comparison between the hierarchical and non-hierarchical (flat) classification tasks for satellite image data is presented in Figure 7.1. hierarchical classification
Early breast cancer diagnostics based on hierarchical machine learning classification for mammography images
Published in Cogent Engineering, 2021
M. Saeed Darweesh, Mostafa Adel, Ahmed Anwar, Omar Farag, Ahmed Kotb, Mohamed Adel, Ayman Tawfik, Hassan Mostafa
In (Setiawan et al., 2015), the authors use the Law’s Texture Energy Measure (LAWS) technique to extract secondary features from the images. First, the Region of Interest (ROI) is extracted using the labels in the dataset, followed by feature extraction using the LAWS feature extraction technique. A hierarchical classification approach is used. Firstly, the classification between normal and abnormal breasts is carried out, followed by another classification between benign and malignant cases if the first stage classification is abnormal. The accuracy, specificity, and sensitivity for classifying normal and abnormal cases are 93.9%, 100%, and 91%, respectively. The accuracy, specificity, and sensitivity of classifying benign and malignant cases are 83.3%, 88%, and 80%, respectively.
Comparing the concept images and hierarchical classification skills of students at different educational levels regarding parallelograms: a cross-sectional study
Published in International Journal of Mathematical Education in Science and Technology, 2022
Definitions in geometry play a central role in understanding the construction of meaning and the essence of geometric concepts, theorems, and proofs (Usiskin et al., 2008). Tall and Vinner (1981) associate students’ missing information and misconceptions of geometric concepts with the disconnection and weak links between the components of ‘concept image’ and ‘concept definition’. According to Tall and Vinner (1981), concept image refers to the whole cognitive structure that includes mental pictures, features, and processes of the concept, while concept definition refers to the set of words used to express the concept. They also highlight that concept definitions and concept images are not always compatible with each other and a concept image is shaped by the knowledge and experiences of the individual, thus thereby intrinsic to that individual. In this sense, de Villiers (1994) mentions two types of classifications of quadrilaterals: hierarchical and partitional (prototype). The hierarchical classification allows for the inclusion of more particular concepts as subsets of more general concepts, whereas in a partitional classification the various subsets of concepts are considered to be disjoint from one another. Similarly, Usiskin et al. (2008) use the term ‘exclusive definition’ instead of ‘partitional classification’ and ‘inclusive definition’ instead of ‘hierarchical classification’. Therefore, while the inclusive definition leads to a hierarchical classification of geometric concepts, the exclusive definition leads to a partitional classification. Thus, we see a relationship between definitions of geometric concepts and classifications of these concepts.