Wyniki wyszukiwania - BazTech

1

Analysis of the possibility of using an audio signal to detect the state of the machine based on the sound it generates

Majda-Zdancewicz Ewelina, Pierzak Wiktor

Przegląd Elektrotechniczny

|

2025

|

R. 101, nr 2

219--223

EN

As part of the experiments, research was carried out to detect the condition of the electric machine. In the first place, a laboratory stand was designed to test acoustic signals. This process included, m.in, the selection of an electric machine, load system, measuring devices, recording device, and the definition of its operating states. The recordings were recorded in the Laboratory of Electrical Machines at the Institute of Electronic Systems of the Faculty of Electronics of the Military University of Technology using the SZUAa54a induction motor. The feature extraction process involved the generation of spectral descriptors. The method of decision trees and the k nearest neighbor method were used as the classification algorithm. The proposed method gives an efficiency of 95.53%.

PL

W ramach eksperymentów przeprowadzono badania mające na celu wykrycie stanu maszyny elektrycznej. W pierwszej kolejności zaprojektowano stanowisko laboratoryjne do badania sygnałów akustycznych. Proces ten obejmował m.in. dobór maszyny elektrycznej, układu obciążeniowego, urządzeń pomiarowych, urządzenia rejestrującego oraz określenie jego stanów pracy. Nagrania rejestrowano w Laboratorium Maszyn Elektrycznych Instytutu Systemów Elektronicznych Wydziału Elektroniki Wojskowej Akademii Technicznej przy użyciu silnika indukcyjnego SZUAa54a. Proces ekstrakcji cech obejmował generację deskryptorów widmowych. Jako algorytm klasyfikacji zastosowano metodę drzew decyzyjnych oraz metodę najbliższego sąsiada. Proponowana metoda daje sprawność 95,53%.

2

Ore image target detection based on improved YOLOv5 network

Cai Gaipin, Luo Hui, Huang Jinruo, Yi Guotao

Gospodarka Surowcami Mineralnymi

|

2025

|

Vol. 41, iss. 2

177--196

EN

The existing target detection algorithms detect the ore on the conveyor belt after the crushing process with low precision and slow detection speed. This leads to challenges in achieving a balance between precision and speed, to enhance the detection precision and speed of ore, and in view of the problems of leakage, misdetection, and insufficient feature extraction of YOLOv5 in the task of ore image detection; this study presents a target detection approach relying on the CA attention mechanism (Coordinate attention for efficient mobile network design), the SIoU loss function and the target detection algorithm YOLOv5 combination of ore image particle target detection method. Integrating the CA attention mechanism into the YOLOv5 backbone feature network enhances the feature learning and extraction of ore images, thereby improving the precision of the detection model; the SIoU loss function is refined to boost the recognition precision of the network on ore images and address the shortcomings of the original loss function that fails to take angular loss, distance loss, and shape loss into account, thereby further improving the precision and speed of ore image detection. The experimental findings demonstrate that the AP value, value, and precision rate are improved compared with the pre-improved algorithm. The CA-YOLOv5 method is verified to be fast, effective, and advanced and provides a foundation for real-time target detection of ores on conveyor belts in subsequent intelligent mine production.

PL

Istniejące algorytmy wykrywania celu wykrywają rudę na taśmie przenośnika po procesie kruszenia z niską precyzją i niską szybkością wykrywania. Prowadzi to do wyzwań związanych z osiągnięciem równowagi między precyzją i szybkością, w celu zwiększenia precyzji i szybkości wykrywania rudy, a także ze względu na problemy z wyciekami, błędnym wykrywaniem i niewystarczającą ekstrakcją cech YOLOv5 w zadaniu wykrywania obrazu rudy; niniejsze badanie przedstawia podejście do wykrywania celu polegające na mechanizmie uwagi CA (Coordinate attention for efficient mobile network design), funkcji straty SIoU i kombinacji algorytmu wykrywania celu YOLOv5 w połączeniu z metodą wykrywania celu cząstek obrazu rudy. Zintegrowanie mechanizmu uwagi CA z siecią funkcji szkieletowych YOLOv5 usprawnia uczenie się funkcji i ekstrakcję obrazów rudy, tym samym zwiększając precyzję modelu wykrywania; funkcja straty SIoU została udoskonalona w celu zwiększenia precyzji rozpoznawania sieci na obrazach rudy i usunięcia niedociągnięć oryginalnej funkcji straty, która nie uwzględnia strat kątowych, strat odległości i strat kształtu, co jeszcze bardziej poprawia precyzję i szybkość wykrywania obrazów rudy. Wyniki eksperymentów pokazują, że wartość AP, wartość i wskaźnik precyzji są lepsze w porównaniu z wcześniej ulepszonym algorytmem. Metoda CA-YOLOv5 została zweryfikowana jako szybka, skuteczna i zaawansowana oraz stanowi podstawę do wykrywania celów rud na taśmach przenośnikowych w czasie rzeczywistym w późniejszej inteligentnej produkcji kopalnianej.

3

Optimized 3D-2D CNN for automatic mineral classification in hyperspectral images

Attallah Youcef, Zigh Ehlem, Adda Ali Pacha

Reports on Geodesy and Geoinformatics

|

2024

|

Vol. 118

82--91

EN

Mineral classification using hyperspectral imaging represents an essential field of research improving the understanding of geological compositions. This study presents an advanced methodology that uses an optimized 3D-2D CNN model for automatic mineral identification and classification. Our approach includes such crucial steps as using the Diagnostic Absorption Band (DAB) selection technique to selectively extract bands that contain the absorption features of minerals for classification in the Cuprite zone. Focusing on the Cuprite dataset, our study successfully identified the following minerals: alunite, calcite, chalcedony, halloysite, kaolinite, montmorillonite, muscovite, and nontronite. The Cuprite dataset results with an overall accuracy rate of 95.73 % underscore the effectiveness of our approach and a significant improvement over the benchmarks established by related studies. Specifically, ASMLP achieved a 94.67 % accuracy rate, followed by 3D CNN at 93.86 %, SAI-MLP at 91.03 %, RNN at 89.09 %, SPE-MLP at 85.53 %, and SAM at 83.31 %. Beyond the precise identification of specific minerals, our methodology proves its versatility for broader applications in hyperspectral image analysis. The optimized 3D-2D CNN model excels in terms of mineral identification and sets a new standard for robust feature extraction and classification.

4

Amodel for processing and identifying engine vibration signals

Nguyen Phuong X., Tran Linh H.

Przegląd Elektrotechniczny

|

2024

|

R. 100, nr 5

263--266

EN

The advancement of the sensor technology becoming increasingly cost-effective and the progress in diagnostic and management research, users nowadays not only demand high reliability from their devices but also the ability for their equipment to self-diagnose errors and provide alerts. These devices often incorporate sensor systems capable of generating plenty of data points, that needed a carefully targeted algorithms for extracting features from the data for classification and prediction models. In this paper, we will develop a comprehensive model for identifying vibration signals. We will extract features from the bearing data provided by Case Western Reserve University Bearing Data Center, then use a deep-learning based convolutional neural network to learn to be a classification model of the motor states based on the vibration signals. The numerical results show that the method can offer the promising accuracy at 85.8%.

PL

Wraz z postępem w technologii czujników, która staje się coraz bardziej tańsza do użycia w badaniach diagnostycznych, użytkownicy wymagają obecnie nie tylko wysokiej niezawodności swoich urządzeń, ale także zdolności ich sprzętu do samodiagnostyki błędów i generowania alertów. Nowej generacje urządzeń zawierają systemy czujników zdolne do generowania mnóstwo danych, co wymagało starannie dobranych algorytmów do wyodrębniania cech charakterystycznych na potrzeby modeli klasyfikacji i predykcji. W tym artykule przedstawimy model do identyfikacji sygnałów drganiowych. Korzystaliśmy z danych pomiarowych łożysk dostępnych w Centrum baz danych łożysk Uniwersytetu Case Western Reserve. Z tych danych pomiarowych, wygenerowano ich spectrogramy do postaci obrazów a następnie wykorzystano splotową sieć neuronową opartą na głębokim uczeniu się do tworzenia model klasyfikacji stanów silnika w oparciu o sygnały wibracyjne. Wyniki liczbowe pokazują, że metoda ta może zapewnić obiecującą dokładność na poziomie 85,8%.

5

Building Material Recognition and Feature Extraction Using Small CCD Sensor and Image Analysis and Clustering Techniques

Pagnotta Stefano, Filimon Danis Ionut, Gallello Gianni, Lezzerini Marco

Inżynieria Mineralna

|

2024

|

Vol. 1, no. 1

229-236

EN

Heritage Building material recognition is the process of classifying building materials based on their visual appearance. It is important in construction, urban planning, and archaeology. Image analysis is a common approach, starting with acquiring RGB images, then extracting features using techniques such as colour histograms and texture analysis, and clustering the materials into groups using algorithms like k-means. Finally, the materials are classified into categories using classifiers like decision trees, SVM, or neural networks. Image analysis is a useful tool for building material recognition, as it allows for accurate classification of building materials based on their visual characteristics.

6

Enhancing multi-class prediction of skin lesions with feature importance assessment

Paulauskaite-Taraseviciene Agne, Sutiene Kristina, Dimsa Nojus, Valiukeviciene Skaidra

International Journal of Applied Mathematics and Computer Science

|

2024

|

Vol. 34, no. 4

617--629

EN

Numerous image processing techniques have been developed for the identification of various types of skin lesions. In real-world scenarios, the specific lesion type is often unknown in advance, leading to a multi-class prediction challenge. The available evidence underscores the importance of employing a comprehensive array of diverse features and subsequently identifying the most important ones as a crucial step in visual diagnostics. For this purpose, we addressed both binary and five-class classification tasks using a small dataset, with skin lesions prevalent in Lithuania. The model was trained using a rich set of 662 features, encompassing both conventional image features and graph-based ones, which were obtained from the superpixel graph generated using Delaunay triangulation. We explored the influence of feature importance determined by SHAP values, resulting in a weighted F1-score of 92.48% for the two-class classification and 71.21% for the five-class prediction.

7

A review of shockable arrhythmia detection of ECG signals using machine and deep learning techniques

Kavya Lakkakula, Karuna Yepuganti, Saritha Saladi, Prakash Allam Jaya, Patro Kiran Kumar, Sahoo Suraj Prakash, Tadeusiewicz Ryszard, Pławiak Paweł

International Journal of Applied Mathematics and Computer Science

|

2024

|

Vol. 34, no. 3

485--511

EN

An electrocardiogram (ECG) is an essential medical tool for analyzing the functioning of the heart. An arrhythmia is a deviation in the shape of the ECG signal from the normal sinus rhythm. Long-term arrhythmias are the primary sources of cardiac disorders. Shockable arrhythmias, a type of life-threatening arrhythmia in cardiac patients, are characterized by disorganized or chaotic electrical activity in the heart’s lower chambers (ventricles), disrupting blood flow throughout the body. This condition may lead to sudden cardiac arrest in most patients. Therefore, detecting and classifying shockable arrhythmias is crucial for prompt defibrillation. In this work, various machine and deep learning algorithms from the literature are analyzed and summarized, which is helpful in automatic classification of shockable arrhythmias. Additionally, the advantages of these methods are compared with existing traditional unsupervised methods. The importance of digital signal processing techniques based on feature extraction, feature selection, and optimization is also discussed at various stages. Finally, available databases, the performance of automated algorithms, limitations, and the scope for future research are analyzed. This review encourages researchers’ interest in this challenging topic and provides a broad overview of its latest developments.

8

Overview of techniques for detecting object’s features and embedding them in multidimensional spaces

Wasiluk Rafał

Computer Science and Mathematical Modelling

|

2024

|

No. 20

29--37

EN

It is evident that each object in the real world possesses unique properties. A subset of these characteristics can be readily described in quantitative terms. Examples of such features include the number of wheels in a vehicle, the floor area of a residential property, or the year of construction of a building. However, certain characteristics of objects exhibit a higher level of complexity. Examples of such features include object shape, color, and texture. These characteristics, frequently defined in terms of objects depicted in images, represent the primary characteristics that can be identified in real-world objects. The processing of these visual attributes has been the subject of scientific research for decades, and the literature on this topic is extensive. The objective of this article is to synthesize the existing methods for detecting object’s shape, color, and texture and embedding them in multidimensional spaces. By applying these methods, it is possible to represent the features of the object as points in multidimensional spaces. Such representations can be used to solve multicriteria optimization problems.

PL

Każdy obiekt w otaczającym nas świecie posiada unikalne cechy. Część z nich, np. ilość kół w pojeździe, powierzchnia nieruchomości lub rok budowy budynku, może być z łatwością opisana przy wykorzystaniu liczb. Istnieją jednak takie cechy obiektów, które charakteryzują się wyższą złożonością i nie można ich w tak prosty sposób opisać. Przykładem takich właściwości może być: kształt, zestaw kolorów lub faktura materiału obiektu. Cechy te są definiowane w kontekście obiektów przedstawionych na obrazach i opisują najbardziej podstawowe właściwości obiektów świata rzeczywistego. Tematyka przetwarzania wspomnianych cech wizualnych jest tematem badań naukowych od wielu dekad, a powstała w tym zakresie literatura jest bardzo obszerna. Niniejszy artykuł zbiera najważniejsze z istniejących podejść do zagadnienia wykrywania kształtu obiektu, jego kolorów, faktury materiału oraz osadzania ich reprezentacji w przestrzeniach wielowymiarowych. Wykorzystanie tych metod czyni możliwym przedstawienie cech obiektu pod postacią punktów w przestrzeniach wielowymiarowych. To z kolei otwiera drogę do wykorzystania przygotowanych reprezentacji w rozwiązywaniu zadań optymalizacji wielokryterialnej.

9

A systematic review of artificial neural network techniques for analysis of foot plantar pressure

Wang Chongguang, Evans Kerrie, Hartley Dean, Morrison Scott, Veidt Martin, Wang Gui

Biocybernetics and Biomedical Engineering

|

2024

|

Vol. 44, no. 1

197--208

EN

Plantar pressure distribution offers insights into foot function, gait mechanics, and foot-related issues. This systematic review presents an analysis of the use of artificial neural network techniques in the context of plantar pressure analysis. 60 studies were included in the review. Sample size, pathology, pressure sensor number, data collection device, utilization of other sensor devices, ground-truth methods, pre-processing dataset, neural network type, and evaluation metrics were evaluated. Utilization of customized wearable footwear devices for the acquisition of data was common amongst both healthy participants and patients. Inertial measurement units emerged as an effective compensatory measure to address the limitations associated with the distribution of plantar pressure. Ground truth methods predominantly relied on the usage of both annotations and reference devices. Multilayer perceptron, convolutional neural networks, and recurrent neural networks were identified as the most frequently employed artificial neural network algorithms across the reviewed studies. Finally, the evaluation of performance largely drew upon statistical descriptions and other machine learning methods. This review provides a comprehensive understanding of the use of artificial neural network techniques in plantar pressure analysis, highlighting opportunities for future research.

10

Impact of Spelling and Editing Correctness on Detection of LLM-Generated Emails

Gryka Paweł, Gradoń Kacper, Kozłowski Marek, Kutyła Miłosz, Janicki Artur

Annals of Computer Science and Information Systems

|

2024

|

Vol. 39

603--608

EN

In this paper, we investigated the impact of spelling and editing correctness on the accuracy of detection if an email was written by a human or if it was generated by a language model. As a dataset, we used a combination of publicly available email datasets with our in-house data, with over 10k emails in total. Then, we generated their “copies'' using large language models (LLMs) with specific prompts. As a classifier, we used random forest, which yielded the best results in previous experiments. For English emails, we found a slight decrease in evaluation metrics if error-related features were excluded. However, for the Polish emails, the differences were more significant, indicating a decline in prediction quality by around 2% relative. The results suggest that the proposed detection method can be equally effective for English even if spelling- and grammar-checking tools are used. As for Polish, to compensate for error-related features, additional measures have to be undertaken.

11

Experimenting with manual and automated data mining pipelines on the FedCSIS 2024 Data Science Challenge

Buck Luisa, Furier Marc, Göktepe Okan Mert, Judák Jusztina, Lautenbach Max, Münker Gregor

Annals of Computer Science and Information Systems

|

2024

|

Vol. 39

751--754

EN

This paper reviews the 5th-best solution and results of the FedCSIS 2024 Data Science Challenge, which aimed to predict stock trends using financial indicators. It details the preprocessing, modelling, and tuning approaches and demonstrates, as well as the methods and techniques used to address the prediction problem effectively. Subsequently, the results of different experiments, including hyperparameter optimization on preprocessing steps and switching between different prediction targets, could be compared to manual experiments. Overall, a manually experienced model could be found to outperform hyperparameter-tuned pipelines.

12

Water agricultural management based on hydrology using machine learning techniques for feature extraction and classification

Lin Yi-Chia, Alorfi Almuhannad Sulaiman, Hasanin Tawfiq, Arumugam Mahendran, Alroobaea Roobaea, Alsafyani Majed, Alghamdi Wael Y.

Acta Geophysica

|

2024

|

Vol. 72, no. 3

1945--1955

EN

For irrigation in agriculture, water is a natural resource. Recycling water use is vital for the sustainable development of ecological environment and for resource conservation. Different substances that are thought to be pollutants and contribute to the deterioration of water quality are present in the wastewater from daily life and industrial activity. This research propose novel method in agricultural water management using feature extraction as well as classification based on DL methods. Inputs are collected as agriculture field water management as well as processed for noise removal, normalization and smoothening. Processed input data features are extracted utilizing kernel convolutional component analysis network. The extracted features has been classified using Quadratic reinforcement NN. Experimental analysis are carried out in terms of accuracy, precision, recall, positive predictive value, RMSE and mAP. Proposed technique attained accuracy of 92%, precision of 86%, recall of 65%, positive predictive value of 71%, RMSE of 55%, MAP of 51%.

13

Aquatic ecosystem-based water management in agriculture project by data analytics using classification by deep learning techniques

Anuradha Tadiparthi, Sen Sanjay Kumar, Tamilarasi Kathirvel Murugan, Haleem Sulaima Lebbe Abdul, Abdul-Samad Zulkiflee, Anupong Wongchai

Acta Geophysica

|

2024

|

Vol. 72, no. 3

2059--2069

EN

Over the past few decades, irrigation using groundwater has increased significantly. It has significant effects on local to regional climates as well as terrestrial energy fluxes, food production, and water availability. High cost of metering equipment installation as well as maintenance, privacy concerns, and existence of unregistered or illegal wells make it difficult to monitor irrigation water use on a large scale. This study suggests a unique approach to DL-based feature extraction and categorization for ecosystem-based water management in agricultural fields. Agriculture field water analysis data were used as the input in this instance, which was subsequently processed for noise removal, smoothing, and normalisation. Particle swarm-based convolutional architecture has been used to extract the processed data feature. Back regressive propagation based on incentive Q-learning is used to classify the extracted features. Experimental analysis has been carried out in terms of accuracy, precision, recall, F-1 score, RMSE and mAPE. Proposed technique obtained accuracy of 92%, precision of 78%, recall of 83%, F_1 score of 76%, RMSE of 55% and MAPE of 57%.

14

Detection of epileptic seizures with the use of convolutional neural networks

Wiszniewski Przemysław, Kołodziej Marcin, Majkowski Andrzej, Rysz Andrzej

Przegląd Elektrotechniczny

|

2023

|

R. 99, nr 2

51--55

EN

The purpose of the article is to investigate whether the implementation of a CNN consisting of several layers will allow the effective detection of epileptic seizures. For the research, a publicly available database registered for 4 dogs and 8 people was used. The 1-second iEEG recordings were marked by a neurophysiologist as interictal, early seizure, and seizure. A CNN was trained for each patient individually. Coefficients such as precision, AUC, sensitivity, and specificity were calculated, and the results were compared with the best algorithms published in one of the contests on the Kaggle platform. The average accuracy for the recognition of seizures using CNN is 0.921, the sensitivity is 0.850, and the specificity is 0.927. For early seizures these values are 0.825, 0.782, and 0.828, respectively.

PL

Celem artykułu było zbadanie czy zastosowanie sieci CNN, składającej się z kilku warstw umożliwi skuteczną detekcję napadów epileptycznych. Na użytek badań zastosowano ogólnodostępną bazę danych zarejestrowaną dla 4 psów oraz 8 ludzi. Jednosekundowe zapisy sygnału iEEG zostały oznaczone przez neurofizjologa jako: międzynapadowe, wczesnonapadowe oraz napadowe. Zaproponowano strukturę sieci CNN, a następnie wytrenowano ją dla każdego pacjenta indywidualnie. Zostały wyliczone współczynniki takie jak: trafność, AUC, czułość, specyficzność. Następnie wyniki zostały porównane do osiągniętych w najlepszych algorytmach opublikowanych w konkursie na platformie Kaggle. Średnia skuteczność rozpoznawania napadów z wykorzystaniem sieci CNN wynosi 0.921, czułość 0.850, a specyficzność 0.927. Dla okresów wczesnonapadowych wartości te wynoszą odpowiednio 0.825, 0.782 i 0.828.

15

Efficient Covid-19 disease diagnosis based on cough signal processing and supervised machine learning

Bensid Khaled, Lati Abdelhai, Benlamoudi Azeddine, Ghouar Brahim Elkhalil, Senoussi Mohammed Larbi

Diagnostyka

|

2023

|

Vol. 24, No. 1

art. no. 2023103

EN

The spread of the coronavirus has claimed the lives of millions worldwide, which led to the emergence of an economic and health crisis at the global level, which prompted many researchers to submit proposals for early diagnosis of the coronavirus to limit its spread. In this work, we propose an automated system to detect COVID-19 based on the cough as one of the most important infection indicators. Several studies have shown that coughing accounts for 65% of the total symptoms of infection. The proposed system is mainly based on three main steps: first, cough signal detection and segmentation; second, cough signal extraction; and third, three techniques of supervised machine learning-based classification: Support Vector Machine (SVM), KNearest Neighbours (KNN), and Decision Tree (DT). Our proposed system showed high performance through good accuracy values, where the best accuracy for classifying female coughs was 99.6% using KNN and 88% for males using SVM.

16

Research on communication emitter identification based on semi-supervised dimensionality reduction in complex electromagnetic environment

Ge Wei, Qi Lin, Tong Lin, Zhu Jun, Zhang Jing, Zhao Dongyang, Li Ke

Bulletin of the Polish Academy of Sciences. Technical Sciences

|

2023

|

Vol. 71, nr 4

art. no. e145766

EN

The individual identification of communication emitters is a process of identifying different emitters based on the radio frequency fingerprint features extracted from the received signals. Due to the inherent non-linearity of the emitter power amplifier, the fingerprints provide distinguishing features for emitter identification. In this study, approximate entropy is introduced into variational mode decomposition, whose features performed in each mode which is decomposed from the reconstructed signal are extracted while the local minimum removal method is used to filter out the noise mode to improve SNR. We proposed a semi-supervised dimensionality reduction method named exponential semi-supervised discriminant analysis in order to reduce the high-dimensional feature vectors of the signals, and LightGBM is applied to build a classifier for communication emitter identification. The experimental results show that the method performs better than the state-of-the-art individual communication emitter identification technology for the steady signal data set of radio stations with the same plant, batch and model.

17

Ensemble of feature extraction methods to improve the structural damage classification in a wind turbine foundation

Leon-Medina Jersson X., Parés Núria, Anaya Maribel, Tibaduiza Diego A., Pozo Francesc

Bulletin of the Polish Academy of Sciences. Technical Sciences

|

2023

|

Vol. 71, nr 3

art. no. e144606

EN

The condition monitoring of offshore wind power plants is an important topic that remains open. This monitoring aims to lower the maintenance cost of these plants. One of the main components of the wind power plant is the wind turbine foundation. This study describes a data-driven structural damage classification methodology applied in a wind turbine foundation. A vibration response was captured in the structure using an accelerometer network. After arranging the obtained data, a feature vector of 58 008 features was obtained. An ensemble approach of feature extraction methods was applied to obtain a new set of features. Principal Component Analysis (PCA) and Laplacian eigenmaps were used as dimensionality reduction methods, each one separately. The union of these new features is used to create a reduced feature matrix. The reduced feature matrix is used as input to train an Extreme Gradient Boosting (XGBoost) machine learning-based classification model. Four different damage scenarios were applied in the structure. Therefore, considering the healthy structure, there were 5 classes in total that were correctly classified. Five-fold cross validation is used to obtain a final classification accuracy. As a result, 100% of classification accuracy was obtained after applying the developed damage classification methodology in a wind-turbine offshore jacket-type foundation benchmark structure.

18

Filtering Decision Rules Driven by Sequential Forward and Backward Selection of Attributes: An Illustrative Example in Stylometric Domain

Zielosko Beata, Stańczyk Urszula, Jabloński Kamil

Annals of Computer Science and Information Systems

|

2023

|

Vol. 35

833--842

EN

The paper presents investigations concerning the decision rule filtering process controlled by the estimated relevance of available attributes. In the conducted study, two search directions were used, sequential forward selection and sequential backward elimination, applied after the knowledge discovery step to the rule sets inferred from a dataset. The steps of sequential search, along with two different strategies of rule selection, were governed by three rankings obtained for variables, all related to characteristics of data and rules that can be induced, as follows, (i) a ranking based on the weighting factor referring to the occurrence of attributes in generated decision reducts, (ii) the OneR ranking exploiting short rule properties, and (iii) the proposed ranking defined through the operation of greedy algorithm for rule induction. The three rankings were confronted and compared from the perspective of their usefulness for the selection of rules performed in the two directions. The resulting sets of rules were analysed with respect to the properties of the constituent decision rules and from the point of performance for all constructed rule-based classifiers. Substantial experiments were carried out in the stylometric domain, treating the task of authorship attribution as classification. The results obtained indicate that for all three rankings and search paths it was possible to obtain a noticeable reduction of attributes while at least maintaining the power of inducers, at the same time improving characteristics of rule sets.

19

Emotion-Based Literature Books Recommender Systems

Luțan Elena-Ruxandra, Bădică Costin

Annals of Computer Science and Information Systems

|

2023

|

Vol. 35

275--280

EN

In this paper we propose two book recommendation methods based on emotions extracted from user reviews, using content-based filtering and collaborative filtering. The methods were experimentally evaluated on our own dataset that we collected from Goodreads -- a popular website with large database of books and readers reviews. We created an experimental setup where the recommendation algorithms for carrying out the evaluation using two proposed evaluation metrics: coverage and average recommendations similarity.

20

Integration of feature extraction, attribute combination and image segmentation for object delineation on seismic images

Khayer Keyvan, Hosseini Fard Esmail, Roshandel Kahoo Amin, Soleimani Monfared Mehrdad, Ahmadyfard Alireza

Acta Geophysica

|

2023

|

Vol. 71, no. 1

275--292

EN

Automatic geological interpretation, specifically modeling salt dome and fault detection, is controversial task on seismic images from complex geological media. In advanced techniques of seismic interpretation and modeling, various strategies are utilized for combination and integration different information layers to obtain an image adequate for automatic extraction of the object from seismic data. Efficiency of the selected feature extraction, data integration and image segmentation methods are the most important parameters that affect accuracy of the final model. Moreover, quality of the seismic data also affects confidence of the selected seismic attributes for integration. The present study proposed a new strategy for efficient delineation and modeling of geological objects on the seismic image. The proposed method consists of extraction specific features by the histogram of oriented gradients (HOG) method, statistical analysis of the HOG features, integration of features through hybrid attribute analysis and image classification or segmentation. The final result is a binary model of the target under investigation. The HOG method here modified accordingly for extraction of the related features for delineation of salt dome and fault zones from seismic data. The extracted HOG parameter then is statically analyzed to define the best state of information integration. The integrated image, which is the hybrid attribute, then is used for image classification, or image segmentation by the image segmentation method. The seismic image labeling procedure performs on the related seismic attributes, evaluated by the extracted HOG feature. Number of HOG feature and the analyzing parameters are also accordingly optimized. The final image classification then is performed on an image which contains all the embedded information on all the related textural conventional and statistical attributes and features. The proposed methods here apply on four seis mic data examples, synthetic model of salt dome and faults and two real data that contain salt dome and fault. Results have shown that the proposed method can more accurately model the targets under investigation, compared to advanced extracted attributes and manual interpretations.