First Step In Classification Research Articles

Новітній підхід до застосування нормалізованого диференційного вегетаційного індексу для класифікування аерокосмічних зображень за наявності неповних та неточних даних

Solution of different scientific, ecological and agricultural tasks with the use of aerospace im-ages comprises a procedure of image classification. Classification is one of the most important procedures. Nowadays many supervised and unsupervised classification methods are applied in remote sensing. The most accurate results are obtained through the use of supervised classifica-tion methods. In this paper, there are proposed some new approaches to image classification which are based on supervised classification methods and Normalized Difference Vegetation Index (NDVI). Different values of NDVI are noted to correspond to different classes of objects, such as soil, water, roads, sand, green vegetation, oil spills. Application of Vegetation Index is the first step of classification. Using NDVI, it is possible to select special necessary classes. Af-ter the application of NDVI, such classification methods as the parallelepiped method, Demp-ster’s rule, and Inagaki’s combination rule can be used. The current work describes the main advantages of these classification methods. It has been noted that the use of the parallelepiped method allows easy and quick processing of data. The paper also shows that Dempster’s combi-nation rule and Inagaki’s combination rule can deal with inaccurate and incomplete data from different spectral bands. Moreover, these methods can process conflicting information. Demp-ster-Shafer theory has the advantage of high accuracy and simple calculations. In the paper, there is also considered a numerical example where NDVI and Inagaki’s combination rule has been used for detection and mapping of oil spills. Application of Vegetation Index and such su-pervised classification rules as the parallelepiped method, Dempster’s rule, Inagaki’s combina-tion rule can be applied in ecological monitoring, mapping of petroleum spills, and solving ag-ricultural tasks.

Depositional environment classification based on environmental variables of South Korea's Nakdong River Estuary barrier-lagoon system

Anthropogenic activities have altered the geomorphological and ecological conditions of the Nakdong River Estuary (NRE) dramatically over the last century. The objectives of this study were to classify NRE sub-environments and to identify their unique ecological functions. The first step in classification was the establishment of 14 a priori sub-environmental groups based on landscape factors. Surface sediments obtained for these groups were analyzed for factors related to grain size and organic matter. Based on the results, the NRE was divided into two primary estuarine environment divisions: (1) a mixed marine and terrestrial environment influenced primarily by land; and (2) a principally marine environment influenced primarily by the ocean. Using multivariate analysis, we subdivided these primary estuarine divisions into six sub-environments, including Sub-1 and -2 in the former and Sub-3, -4, -5, and -6 in the latter.

Precision Medicine: The new era in medicine

Precision Medicine: The new era in medicine

A single cataluminescence sensor for wine identification by luminescent response profiles

In this paper, a new method involving cataluminescence (CTL) response profiles combined with chemometrics was developed in an attempt to classify wine of different types (health liquor, brandy, glutinous wine, beer and grape wine). In a closed reaction cell (CRC), with an optimal injection volume of 10 μl, reaction temperature of 350 °C and detection wavelength of 425 nm, the unique response profiles of wine were obtained on the surface of nano-ZrO2. The comparison of the characteristic profiles with marked extreme points, as the first step of classification, showed that the extreme points of wine of different types were substantially different. The application of linear discriminant analysis (LDA), as the second step of classification, showed that all of the samples were well-discriminated with an accuracy of 100%. The method is effective and fast and does not require sample preparation. The stable and reversible response of this proposed sensor, combined with its simple instrumentation, indicates its promise in real-world wine identification.

Construction of an automated screening system to predict breast cancer diagnosis and prognosis

Background and aim: Using machine learning methods can be helpful in the clinical decision processes such as pathological diagnosis with the aid of microscopic feature datasets. In the present study using the Breast Cancer Wisconsin dataset, an optimal algorithm (classifiers) which can predict both diagnosis (benign vs malignant) and prognosis (recur vs non-recur) was devised by comparing several classification algorithms. Methods: The performance of a two-step algorithm, which sequentially decides diagnosis and prognosis, was compared with that of a multi-class classifier, which divides classes simultaneously. Results: In the two-step classifier, it was discovered that the functional trees (FT) algorithm is the best for the first step of classification, and Naïve Bayes is the best for the second step of classification. On the other hand, the one-step classifier shows better accuracy and better prediction on benign and non-recurring cases than the two-step classifier, but it shows lower accuracy on predicting recurring cases, leading to lower sensitivity. Conclusions: We conclude that the two-step classifier with FT and Naïve Bayes is better than the one-step classifier. This work will be helpful in setting the automated screening system in real clinics and highlight clues to improve the accuracy by refining data and algorithm selection in data mining or machine learning processes.

Classification and trend analysis of UML books (1997–2009)

Technical books of each subject area denote the level of maturity and knowledge demand in that area. According to the Google Books database, about 208 Unified Modeling Language (UML) books have been published from its inception in 1997 until 2009. While various book reviews are frequently published in various sources (e.g., IEEE Software Bookshelf), there are no studies to classify UML books into meaningful categories. Such a classification can help researchers in the area to identify trends and also reveal the level of activity in each sub-area of UML. The statistical survey reported in this article intends to be a first step in classification and trend analysis of the UML books published from 1997 to 2009. The study also sheds light on the quantity of books published in different focus areas (e.g., UML's core concepts, patterns, tool support, Object Constraint Language and Model-Driven Architecture) and also on different application domains (e.g., database modeling, web applications, and real-time systems). The trends of book publications in each sub-area of UML are also used to track the level of maturity, to identify possible Hype cycles and also to measure knowledge demand in each area.

Information gathering for CLP classification.

Regulation 1272/2008 includes provisions for two types of classification: harmonised classification and self-classification. The harmonised classification of substances is decided at Community level and a list of harmonised classifications is included in the Annex VI of the classification, labelling and packaging Regulation (CLP). If a chemical substance is not included in the harmonised classification list it must be self-classified, based on available information, according to the requirements of Annex I of the CLP Regulation. CLP appoints that the harmonised classification will be performed for carcinogenic, mutagenic or toxic to reproduction substances (CMR substances) and for respiratory sensitisers category 1 and for other hazard classes on a case-by-case basis. The first step of classification is the gathering of available and relevant information. This paper presents the procedure for gathering information and to obtain data. The data quality is also discussed.

Fuzzy conceptual clustering

AbstractGrouping unknown data into groups of similar data is a necessary first step for classification, indexing of databases, and prediction. Most of the current applications, such as news classification, blog indexing, image classification, and medical diagnosis, obtain their data in temporal sequence or online. The necessity for data exploration requires a graphical method that allows the expert in the field to study the determined groups of data. Therefore, incremental hierarchical clustering methods that can create explicit cluster descriptions are convenient. The noisy and uncertain nature of the data makes it necessary to develop fuzzy clustering methods. We propose a novel fuzzy conceptual clustering algorithm. We describe the fuzzy objective function for incremental building of the clusters and the relation among the clusters in a hierarchy. The operations that can incrementally reoptimize the fuzzy‐based hierarchy based on the newly arrived data are explained. Finally, we evaluate our method and present the results. The evaluation of the discovered concepts based on a decision tree classifier shows that the accuracy of the decision tree is very good for the fuzzy conceptual clustering algorithm compared with fuzzy c‐means and the accuracy comes close to the expert's accuracy. Copyright © 2010 John Wiley & Sons, Ltd.

PS-31-9 Sympathetic skin response in the face with magnetic stimulation in normal subjects and patients with Parkinson's disease

Cholinergic urticaria (CU) has clinically characteristic features, and has been frequently described in the literature. However, despite its comparatively old history, the pathogenesis and classification remains to be clarified. CU patients are occasionally complicated by anhidrosis and/or hypohidrosis. This reduced-sweat type should be included in the classification because the therapeutic approaches are different from the ordinary CU. It is also well-known that autologous sweat is involved in the occurrence of CU. More than half of CU patients may have sweat hypersensitivity. We attempt to classify CU and address the underlying mechanisms of CU based on the published data and our findings. The first step for classification of CU seems to discriminate the presence or absence of hypersensitivity to autologous sweat. The second step is proposed to determine whether the patients can sweat normally or not. With these data, the patients could be categorized into three subtypes: (1) Cu with sweat hypersensitivity; (2) CU with acquired anhidrosis and/or hypohidrosis; (3) idiopathic CU. The pathogenesis of each subtype is also discussed in this review.

自己組織化ニューラルネットワークの開発と鉱物同定への適用

Self-Organizing Neural Network (SONN) was constructed for the purpose of mineral identification. This system consists of two different kinds of networks, Kohonen's Self-Organizing Map and three layer feedforward neural network based on the back-propagation learning algorithm. The former step, Self-Organizing Map, could divide minerals into some categories by the similarities on the selected characteristics of minerals. This rough division of whole input patterns on feature maps was closely analogous to the first step of classification by human brains. The later step, each category had the three layer feedforward neural network independently, and then the minerals belonging to the same category could be identified.In this study, 82 minerals were identified by 5 characteristics of cleavage, metallic luster, Mohs hardness, streak, and color. Some minerals have plural input patterns on the 5 characteristics mentioned above. Therefore, total number of input patterns was 119 for 82 minerals.After constructing the feature maps and the back-propagation learning, this system could suggest the suitable mineral name for unlearning input patterns. The advantage of the proposed method is that scaling up of the system is possible with relatively small increase in learning times. Further, it should be stressed that this technique can be used in other problems where recognition and identification are necessary.