This paper presents the results of applying the unsupervised learning method (K-means clustering) on the gravity anomaly field in the central region of Vietnam to separate the research area into different clusters, which are homologous in physical properties. In order to achieve the optimal results, the input parameter plays an important role. In this paper, we chose 04 input attributes including the gravity anomalous field attribute, the horizontal gradient attribute, the variance attribute, and the tracing coefficient of the gravity anomalous axis. The obtained results have shown that the research area could be divided into 7 clusters, 9 clusters, 11 clusters, and 13 clusters with close characteristics of the physical properties of the gravity field. The research results show that the Southwest, the Center, and the South of the study area have complex changing physical properties, this result reflects the complicated tectonic activities in these areas with the presence of crumpled and fractured rock layers in different directions and these locations are the potential places to form endogenous mineral deposits of magma origin. The Northwest, the North, and the East parts of the research area witness negligible changes in the field's physical properties, reflecting the stability of the soil and rock layers in this area, with the direction of extending structure from the Northwest to the Southeast. The clustering results according to the K-means unsupervised learning algorithm in central Vietnam initially increase the reliability of the decisions of geologists and geophysicists in interpreting the geological structure and evaluating the origin of deep-hidden mineral deposits in the area.
2
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
This study was conducted to prepare a food susceptibility map in northwest of Hamadan Province, Iran. For this purpose, six criteria related to food (i.e., distance to discharge channel, slope (%), elevation, soil texture and land use, topographic wet index, and check dams) were chosen. Then, based on the role of these criteria on degree of food susceptibility, were weighted both in the context of inter-weighting (fuzzy logic) and outer-criteria (Interval Rough Analytic Hierarchy Process). Finally, by combining these primary weights by weight overlay method in GIS, the food susceptibility mapping was prepared in the study area. The resulted map based on K-means clustering and Silhouette function was divided into 9 clusters, whereas the lower clusters show low susceptibility to food and vice versa. To assess the accuracy of the produced map, 102 food observation points were overlaid on the clustered food susceptibility map. The results showed that among these 102 food points, 66 points are located in the clusters 8 and 9 and 3 points are located on cluster 7. These values show that the produced food susceptibility mapping has a high accuracy.
Big data research has become an important discipline in information systems research. However, the flood of data being generated on the Internet is increasingly unstructured and non-numeric in the form of images and texts. Thus, research indicates that there is an increasing need to develop more efficient algorithms for treating mixed data in big data for effective decision making. In this paper, we apply the classical K-means algorithm to both numeric and categorical attributes in big data platforms. We first present an algorithm that handles the problem of mixed data. We then use big data platforms to implement the algorithm, demonstrating its functionalities by applying the algorithm in a detailed case study. This provides us with a solid basis for performing more targeted profiling for decision making and research using big data. Consequently, the decision makers will be able to treat mixed data, numerical and categorical data, to explain and predict phenomena in the big data ecosystem. Our research includes a detailed end-to-end case study that presents an implementation of the suggested procedure. This demonstrates its capabilities and the advantages that allow it to improve the decision-making process by targeting organizations’ business requirements to a specific cluster[s]/profiles[s] based on the enhancement outcomes.
4
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
Detection of ischemic stroke lesions plays a vital role in the assessment of stroke treatments such as thrombolytic therapy and embolectomy. Manual detection and quantification of stroke lesions is a time-consuming and cumbersome process. In this paper, we present a novel automatic method to detect acute ischemic stroke lesions from Magnetic Resonance Image (MRI) volumes using textural and unsupervised learned features. The proposed method proficiently exploits the 3D contextual evidence using a patch-based approach, which extracts patches randomly from the input MR volumes. Textural feature extraction (TFE) using Gray Level Co-occurrence Matrix (GLCM) and unsupervised feature learning (UFL) based on k-means clustering approaches are employed independently to extract features from the input patches. These features obtained from the two feature extractors are then given as input to the Random Forest (RF) classifier to discriminate between normal and lesion classes. A hybrid approach based on the combination of TFE using GLCM and UFL based on the k-means clustering is proposed in this work. Hybrid combination approach results in more discriminative feature set compared with the traditional approaches. The proposed method has been evaluated on the Ischemic Stroke Lesion Segmentation (ISLES) 2015 training dataset. The proposed method achieved an overall dice coefficient (DC) of 0.886, precision of 0.979, recall of 0.831 and accuracy of 0.8201. Quantitative measures show that the proposed approach is 28.4%, 27.14%, and 5.19% higher than the existing methods in terms of DC, precision, and recall, respectively.
The paper presents an exemplification of data mining techniques in aviation industry on the basis of Turkish Airlines. The purpose of the paper is to present application of data mining on the selected operational data, concerning international flight passenger baggage data, in year 2015. The differences in passenger and flight profiles have been examined. Firstly, two-steps approach allowed defining the number of clusters. Secondly, K-means clustering were applied to divide data into a certain number of clusters representing the different areas of consumption. Results can contribute to higher efficiency in decision making regarding destination offer and fleet management.
6
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
W artykule zaprezentowano autorską metodę detekcji krótkich fragmentów sygnału EEG, które zawierają artefakty mrugania oczami. Autorzy, do automatycznego wskazania fragmentów sygnału EEG zawierającego artefakty mrugania oczami wykorzystali uczenie bez nadzoru (algorytm K-means) oraz cechy sygnału takie jak amplituda i statystyki wyższych rzędów. Wyniki działania algorytmu są bardzo zadowalające. Trafność detekcji wynosi 98%. Algorytm pozwala wykluczyć zaznaczone fragmenty sygnału i nie poddawać ich dalszej analizie. Takie podejście zdaniem autorów przysłuży się do efektywniejszego wykorzystania sygnałów EEG.
EN
The paper presents an original method for the detection of short fragments of the EEG signal, which contain eye blinking artifacts. The authors, to automatically identify fragments the EEG signal containing eye blinking artifacts, used unsupervised learning (K-means algorithm) and the signal features such as amplitude and higher-order statistics. The obtained results are very satisfactory. Accuracy of detection is 98%. The algorithm enables to exclude selected fragments of the signal and not analyze them further. Such an approach, according to the authors, enable more efficient use of EEG signals.
Artykuł przedstawia nową metodę segmentacji sekwencji obrazów termicznych wyodrębniającą obszary o różnych właściwościach cieplnych. Metoda oparta jest na korelacji położenia i kształtu segmentów w poszczególnych kadrach sekwencji. Segmentacja pozwala zmniejszyć liczbę analizowanych obszarów do kilku tysięcy razy, co stwarza realne możliwości praktycznego wykorzystania tomografii termicznej. Opisana metoda jest porównana z algorytmami klasteryzacji K-Means i FCM. Zaletą algorytmu korelacyjnego jest automatyczne wyznaczanie liczby segmentów wyjściowych.
EN
This paper presents a new method for segmentation of thermal image sequences. Its aim is to divide the sequence into segments with different thermal properties. The described algorithm is based on measurements of the position and shape correlation of the segments in successive frames of the sequence. It is composed of several stages. The first stage consists of segmenting consecutive frames of the sequence (Fig. 2). The second step is analysis of the similarity of each segment in each frame with respect to all other segments of all frames and synthesis of the intermediate segments (Fig. 4). The intermediate segments form the segmented output image using the depth buffer technique to resolve multiple pixel-to-segment assignments (Fig. 6). This method is a basis for the thermal analysis of solids, which results in discovering depth profiles of thermal properties for each area. The segmentation reduces the number of the analyzed areas down to a few thousand times, which creates real opportunities for practical application of thermal tomography. The new algorithm has been compared with the K means algorithm [2], and FCM [6], which minimizes the sum of pixel value deviations from the centers of the segments they are assigned to, for all frames of the sequence (Tab. 1). The advantage of the correlation method is automatic determination of the number of output segments in the image and maintaining the constant segmentation error when increasing the number of the processed frames.
Clustering is widely used to explore and understand large collections of data. K-means clustering method is one of the most popular approaches due to its ease of use and simplicity to implement. This paper introduces Density-based Split- and -Merge K-means clustering Algorithm (DSMK-means), which is developed to address stability problems of standard K-means clustering algorithm, and to improve the performance of clustering when dealing with datasets that contain clusters with different complex shapes and noise or outliers. Based on a set of many experiments, this paper concluded that developed algorithms “DSMK-means” are more capable of finding high accuracy results compared with other algorithms especially as they can process datasets containing clusters with different shapes, densities, or those with outliers and noise.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.