ON THE FORMATION OF SPACE FOR THE DESCRIPTION OF THEMATIC DOCUMENTS

2021 ◽  
pp. 35-50
Author(s):  
Ulugbek Yu. Tuliev ◽  
◽  
Musulmon Ya. Lolaev ◽  
Keyword(s):  
Pattern Recognition ◽  
Binary Relation ◽  
Quantitative Estimation ◽  
Feature Space ◽  
Subject Area ◽  
Topological Properties ◽  
Lagrange Method ◽  
Object Property ◽  
Selection Algorithms ◽  
Conditional Optimization

Reducing the dimension of the feature space for describing thematic documents is considered. Descriptions of documents are presented in the form of an “object-property” table, for the formation of which thematic dictionaries were developed with a volume of no more than 100 keywords for each subject area. The correctness of the formation of dictionaries is proved in the framework of the problem of the pattern recognition with disjoint classes. Results of the analysis of the topological properties of the feature space by the values of the compactness measures are used as a research tool. The values of the compactness measures are the quantitative estimation of structures in relations between objects for each class and for the sample as a whole. The structure of relationships is investigated through the division of the class objects into disjoint groups. A path always may be created based on binary relation of connectedness between any two objects of a group. The choice of the space for the description of documents is made by solving the problem of conditional optimization using the Lagrange method. The condition for the formation of an ordered sequence of features is determined. Applying of an ordered sequence is considered as a method to reduce the combinatorial complexity of the selection algorithms. When removing uninformative features from the description of documents, the value of the measure of the compactness of the sample reaches its maximum. A visual representation of the complexity of the configuration of groups and the connectivity of objects from their composition is given.

scholarly journals Land Cover Change Detection Using Object-Based Classification Technique: A Case Study Along The Kosi River, Bihar

2014 ◽  
Vol XL-8 ◽  
pp. 839-843 ◽  
Author(s):  
M. Modi ◽  
R. Kumar ◽  
G. Ravi Shankar ◽  
T.R. Martha
Keyword(s):  
Land Use ◽  
Land Cover ◽  
Time Series Data ◽  
Feature Space ◽  
Series Data ◽  
Land Use Land Cover ◽  
Shape Context ◽  
Object Property ◽  
Object Based ◽  
Remote Sensing Techniques

Land use/land cover (LULC) is dynamic in nature and can affect the ability of land to sustain human activities. The Indo-Gangetic plains of north Bihar in eastern India are prone to floods, which have a significant impact on land use / land cover, particularly agricultural lands and settlement areas. Satellite remote sensing techniques allow generating reliable and near-realtime information of LULC and have the potential to monitor these changes due to periodic flood. Automated methods such as object-based techniques have better potential to highlight changes through time series data analysis in comparison to pixel-based methods, since the former provides an opportunity to apply shape, context criteria in addition to spectral criteria to accurately characterise the changes. In this study, part of Kosi river flood plains in Supaul district, Bihar has been analysed to identify changes due to a flooding event in 2008. Object samples were collected from the post-flood image for a nearest neighbourhood (NN) classification in an object-based environment. Collection of sample were partially supported by the existing 2004–05 database. The feature space optimisation procedure was adopted to calculate an optimum feature combination (i.e. object property) that can provide highest classification accuracy. In the study, for classification of post-flood image, best class separation was obtained by using distance of 0.533 for 28 parameters out of 34. Results show that the Kosi flood has resulted in formation of sandy riverine areas.

scholarly journals A New Method for Biostatistical miRNA Pattern Recognition with Topological Properties of Visibility Graphs in 3D Space

2019 ◽  
Vol 2019 ◽  
pp. 1-9
Author(s):  
Matej Babič ◽  
Ninoslav Marina ◽  
Andrej Mrvar ◽  
Kumar Dookhitram ◽  
Michele Calì
Keyword(s):  
Pattern Recognition ◽  
Biomedical Engineering ◽  
New Method ◽  
Topological Properties ◽  
3D Space ◽  
Visibility Graphs ◽  
Visibility Computation ◽  
Neural Network System ◽  
Dna Pattern ◽  
Visibility Determination

Visibility is a very important topic in computer graphics and especially in calculations of global illumination. Visibility determination, the process of deciding which surface can be seen from a certain point, has also problematic applications in biomedical engineering. The problem of visibility computation with mathematical tools can be presented as a visibility network. Instead of utilizing a 2D visibility network or graphs whose construction is well known, in this paper, a new method for the construction of 3D visibility graphs will be proposed. Drawing graphs as nodes connected by links in a 3D space is visually compelling but computationally difficult. Thus, the construction of 3D visibility graphs is highly complex and requires professional computers or supercomputers. A new method for optimizing the algorithm visibility network in a 3D space and a new method for quantifying the complexity of a network in DNA pattern recognition in biomedical engineering have been developed. Statistical methods have been used to calculate the topological properties of a visibility graph in pattern recognition. A new n-hyper hybrid method is also used for combining an intelligent neural network system for DNA pattern recognition with the topological properties of visibility networks of a 3D space and for evaluating its prospective use in the prediction of cancer.

On normality of a closure space generated from a tree by relation

2021 ◽  
pp. 2250110
Author(s):  
A. K. Das ◽  
Ria Gupta
Keyword(s):  
Comparative Study ◽  
Binary Relation ◽  
Applied Mathematics ◽  
Closure Space ◽  
Topological Properties ◽  
Closure Spaces

Binary relation plays a prominent role in the study of mathematics in particular applied mathematics. Recently, some authors generated closure spaces through relation and made a comparative study of topological properties in the space by varying the property on the relation. In this paper, we have studied closure spaces generated from a tree through binary relation and observed that under certain situation the space generated from a tree is normal.

The Research of the Rolling Element Bearing Fault Signal Frequency Domain Feature Extraction Method Based on the Wavelet Packet Decomposition

2014 ◽  
Vol 937 ◽  
pp. 351-356 ◽  
Author(s):  
Shi Yin Qiu ◽  
Rui Bo Yuan
Keyword(s):  
Pattern Recognition ◽  
Feature Extraction ◽  
Euclidean Distance ◽  
Wavelet Packet ◽  
Feature Space ◽  
Rolling Element Bearing ◽  
Signal Spectrum ◽  
Wavelet Packet Decomposition ◽  
Bearing Fault ◽  
Optimal Feature

The wavelet packet decomposition can be used to extract the frequency band containing bearing fault feature, because the fault signal can be decomposed into different frequency bands by using the wavelet packet decomposition, that is to say the optimal wavelet packet decomposition node needs to be found. A method applying the average Euclidean distance to find the optimal wavelet packet decomposition node was presented. First of all, the bearing fault signals were decomposed into three layers wavelet coefficients by which the bearing fault signals were reconstructed. The peak values extracted from the reconstructing signal spectrum constructed a feature space. Then, the minimum average Euclidean distance calculated from the feature space indicated the optimal wavelet packet node. The optimal feature space could be constructed by the feature points extracted from the signals reconstructed by the optimal wavelet packet nodes. Finally, the optimal feature space was used for the K-means clustering. The feature extraction and pattern recognition test of the four kinds of bearing conditions under four kinds of rotation speeds was detailed. The test results show this method, which can extract the bearing fault feature efficiently and make the fault feature space have the lowest within-class scatter, wons a high pattern recognition accuracy.

scholarly journals From Complex System Analysis to Pattern Recognition: Experimental Assessment of an Unsupervised Feature Extraction Method Based on the Relevance Index Metrics

Computation ◽  
2019 ◽  
Vol 7 (3) ◽  
pp. 39 ◽  
Author(s):  
Laura Sani ◽  
Riccardo Pecori ◽  
Monica Mordonini ◽  
Stefano Cagnoni
Keyword(s):  
Pattern Recognition ◽  
Feature Extraction ◽  
Extraction Method ◽  
System Analysis ◽  
Feature Space ◽  
Extraction Methods ◽  
Feature Extraction Method ◽  
Real World Datasets ◽  
Relevance Index ◽  
Unsupervised Feature Extraction

The so-called Relevance Index (RI) metrics are a set of recently-introduced indicators based on information theory principles that can be used to analyze complex systems by detecting the main interacting structures within them. Such structures can be described as subsets of the variables which describe the system status that are strongly statistically correlated with one another and mostly independent of the rest of the system. The goal of the work described in this paper is to apply the same principles to pattern recognition and check whether the RI metrics can also identify, in a high-dimensional feature space, attribute subsets from which it is possible to build new features which can be effectively used for classification. Preliminary results indicating that this is possible have been obtained using the RI metrics in a supervised way, i.e., by separately applying such metrics to homogeneous datasets comprising data instances which all belong to the same class, and iterating the procedure over all possible classes taken into consideration. In this work, we checked whether this would also be possible in a totally unsupervised way, i.e., by considering all data available at the same time, independently of the class to which they belong, under the hypothesis that the peculiarities of the variable sets that the RI metrics can identify correspond to the peculiarities by which data belonging to a certain class are distinguishable from data belonging to different classes. The results we obtained in experiments made with some publicly available real-world datasets show that, especially when coupled to tree-based classifiers, the performance of an RI metrics-based unsupervised feature extraction method can be comparable to or better than other classical supervised or unsupervised feature selection or extraction methods.

Sign in / Sign up

Export Citation Format

Share Document