Classification Test Research Articles

Bearing fault diagnosis in induction motor using continuous wavelet transform and convolutional neural networks

Induction motors are widely used in various industries due to their high efficiency, reliability and low cost. However, faults in these motors can lead to serious problems, such as unexpected shutdowns, decreased efficiency, and even damage to other parts of the system. Monitoring and diagnosing these faults are necessary. In this study, we propose a new approach for diagnosing bearing faults using convolutional neural network (CNN) and continuous wavelet transform (CWT). The suggested approach uses Scalograms with various CWT types as the network's input and utilizes many epochs and various batch sizes (Multi Ep-Batch) throughout the bearing fault classification training and testing phases. To assess our method, we implemented an extension of the Squeeze Net pre-trained model (transfer learning). The results show that the proposed method outperforms traditional methods in terms of accuracy and computational efficiency in detecting bearing faults. These results are based on publicly available MFPT data, and the proposed approach is compared to traditional methods. This work opens new research avenues in the field of bearing fault diagnosis and provides a promising solution for real-world applications.

Plausible Reasoning in an Algorithm for Generation of Good Classification Tests

Статья посвящена применению принципов (правил) правдоподобных рассуждений к символьному машинному обучению (МО). Эти применения существенны и необходимы для увеличения эффективности алгоритмов МО. Множество таких алгоритмов порождают и используют правила в форме импликаций. Обсуждается генерация этих правил по отношению к классам объектов. Эти классификационные правила специфичны. Их посылки, называемые хорошими замкнутыми тестами (ХЗТ), покрывают максимально возможное множество объектов. Представлен один из алгоритмов генерации ХЗТ, называемый NIAGARA. Алгоритм пересмотрен и предложены новые процедуры на основе правдоподобных рассуждений. Их корректность доказывается. Используются следующие правила: импликации, запреты, индуктивные правила расширения текущих множеств целевых объектов, правила сокращения области поиска решений. Они позволяют увеличить эффективность алгоритма.

Plausible Reasoning in an Algorithm for Generation of Good Classification Tests

Plausible Reasoning in an Algorithm for Generation of Good Classification Tests

Plausible Reasoning in an Algorithm for Generation of Good Classification Tests

The paper is devoted to the application of the plausible reasoning principles to symbolic machine learning. It seems for us that the applications are essential and necessary to improve the efficiency of ML algorithms. Many such algorithms produce and use rules in the form of implication. The generation of these rules with respect to the object classes is discussed. Our classification rules are specific. Their premise part, called good closed tests (GCTs), should cover as many objects as possible. One of the algorithms of GCTs generation called NIAGARA is presented. The algorithm is revisited and new procedures based on plausible reasoning are proposed. Their correctness is proved in propositions. We use the following rules: implication, interdiction, inductive rules of extending current sets of goal-oriented objects, rules of pruning the domain of searching solution. They allow to rise the effectiveness of algorithms.

A new parameter-free entropy based on fragment oscillation and its application in fault diagnosis

For finite time series, it is difficult to provide a convincing determination scheme for specifying relevant parameters of entropy. This article proposes a new entropy, σkEn, based on the amplitude fluctuations of time series segments. It uses the optimal variance estimation under the principle of minimizing AMSE(σˆ2) to adaptively select the window width, and then realizes the no-argument calculation process of volatility entropy through kernel density estimation. By testing with artificial data and applying to real-world data, it has been verified anti-noise and stability properties of σkEn. In a background noise environment with an intensity of 20%, σkEn distinguish various chaotic systems and random signals efficiently. In addition, this manuscript also proposes the plane of (σkEn,NkEn) and (σ̄,σkEn) , which also successfully identifies 6 types of chaotic systems, and 3 random noise signals. Furthermore, (<σ̄>,<σkEn>) curve exhibits different dynamical behavior patterns for these chaotic systems. Finally, we employ (σ̄,σkEn) as a feature and take 8 usual machine learning models, such as Linear Discriminant, Gaussian Naive Bayes and Subspace Discriminant, etc, to perform anomaly detection and fault classification tests on 6 open datasets. Given the appropriate classifiers, our method can achieve a classification accuracy of 100% for four tasks, and an average accuracy of over 96% for the remaining two. This is significantly superior than the accuracy obtained by using Permutation Entropy as a feature.

A comparison of the performance of 68Ga-Pentixafor PET/CT versus adrenal vein sampling for subtype diagnosis in primary aldosteronism.

To investigate the diagnostic efficiency and prognostic value of 68Ga-Pentixafor PET/CT in comparison with adrenal vein sampling (AVS) for functional lateralization in primary aldosteronism (PA). Histology and long-term clinical follow-up normally serve as the gold standard for such diagnosis. We prospectively recruited 26 patients diagnosed with PA. All patients underwent 68Ga-Pentixafor PET/CT and AVS. Postsurgical biochemical and clinical outcomes of patients with unilateral primary aldosteronism (UPA), as diagnosed by PET/CT or AVS, were assessed by applying standardized Primary Aldosteronism Surgical Outcome (PASO) criteria. Immunohistochemistry (IHC) was performed to detect the expression of aldosterone synthase (CYP11B2) and CXCR4. On total, 19 patients were diagnosed with UPA; of these, 13 patients were lateralized by both PET/CT and AVS, four patients were lateralized by PET-only, and two by AVS-only. Seven subjects with no lateralization on AVS and PET received medical therapy. All patients achieved complete biochemical success except one with nodular hyperplasia lateralized by AVS alone. The consistency between PET/CT and AVS outcomes was 77% (20/26). Moreover, CYP11B2-positive nodules were all CXCR4-positive and showed positive findings on PET. Patients who achieved complete biochemical and clinical success had a higher uptake on PET as well as stronger expression levels of CXCR4 and CYP11B2. Our analysis showed that 68Ga-Pentixafor PET/CT could enable non-invasive diagnosis in most patients with PA and identify additional cases of unilateral and surgically curable PA which could not be classified by AVS. 68Ga-Pentixafor PET/CT should be considered as a first-line test for the future classification of PA.

Extreme Learning Machine Classification Method based on Twin Strategy Salp Swarm Algorithm

Extreme Learning Machines (ELM) are a type of Single Hidden Layer Feedforward Neural Network (SLFN). In recent years, they have attracted attention for their powerful approximation ability and fast learning speed. Compared with traditional neural network algorithms, ELM have the advantages of simple structure, fast learning speed, and good generalization performance. However, since the input weights and biases of ELM are randomly generated, there may be some suboptimal or unnecessary input weights and biases. In addition, ELM may require more hidden nodes, which may slow down its response to unknown test data. To address these issues, a twin strategy Salp Swarm Algorithm (TSSA) is proposed to increase the population diversity to improve the convergence speed of the algorithm through chaotic initialization, while shock inertia weights and learning paradigms are introduced into the follower position updating to enhance the stochasticity of the particles. Classification tests are conducted on six datasets using the proposed TSSA, and the experimental results show that the proposed TSSA-ELM has higher accuracy and better generalization performance than some existing ELM variants.

Classification of Rice Growth Phases Using the K-Nearest Neighbor Algorithm in the Irrigation area of Seulimeum Sub District, Aceh, Indonesia

Rice (Oryza sativa L.) is the main food commodity for most of Indonesia’s population. The existence of irrigation areas dramatically affects the water demand in each phase of rice growth. Due to a lack of remote sensing knowledge, classifying the rice growth phase is done manually. This study aims to classify rice growth phases using the K-Nearest Neighbor Algorithm in the irrigation area of the Seulimuem Sub-District. This research uses Google Earth satellite imagery in November 2021 through three stages: multi-resolution segmentation, guided classification, and accuracy testing. The results showed that the paddy and non-paddy fields are 1,425.56 ha and 6,581.70 ha, respectively. The classification results of rice plant growth phases consist of 4 phases, namely the inundation phase, the vegetative phase, the generative phase, and the fallow phase. Based on the accuracy assessment results in November 2021, the overall accuracy ranged from 90.65% to 92.92%, with a kappa index value of 0.85 to 0.87, categorizing it as “almost perfect.”

Characterization of Soil Properties Influencing Flash Flood-Related Road Damage in North-Eastern Regions of Bangladesh

Flash floods in the North-Eastern region of Bangladesh pose a severe threat to road infrastructure, causing extensive damage and disruptions. This study focuses on the comprehensive characterization of soil properties that play a pivotal role in exacerbating flash flood-related road damage. The North-Eastern region, characterized by diverse geological formations, including the Surma Basin and the Meghalaya Plateau, experiences recurrent flash floods during the monsoon season. The undulating topography and a network of rivers contribute to the complexity of the hydrogeological system, influencing soil properties and their impact on road infrastructure.&#x0D; Field surveys were conducted to identify flash flood-prone areas, focusing on elevation, slope, and proximity to water bodies. Soil samples were collected at various depths and locations within vulnerable road sections, considering different land uses and topographical features. Laboratory analyses included particle size distribution, moisture content determination, soil classification, and cohesion tests. The results of this study highlight specific road sections vulnerable to flash floods, providing a foundation for targeted soil characterization. Key soil properties influencing road damage include high clay content, poor drainage characteristics, and low cohesion. The results also revealed the correlation between vulnerable road sections and specific combinations of soil properties, allowing for the identification of hotspots prone to flash flood-related damage.&#x0D; Finally, this study contributes to an enhanced understanding of the soil properties influencing flash flood-related road damage in the North-Eastern region. The identified key soil characteristics offer valuable insights for road infrastructure planning and design, emphasizing the need for mitigation measures such as improved drainage systems and the use of appropriate roadbed materials to enhance overall road resilience in flood-prone areas.

Development of Fishcake Gripping and Classification Automation Process Based on Suction Shape Transformation Gripper

The surge in demand for automating seafood processing necessitates the development of robotic processes for transportation, packaging, and classification. South Korean companies are actively constructing diverse robots and grippers for fishcake handling, yet small workshops face spatial constraints. To address this, the study focuses on creating a gripper capable of versatile fishcake handling within compact spaces. The gripper, designed for single-robot use, employs three suction cups, adapting its grip based on fishcake shapes. Small fishcakes are gripped at the center with one suction cup, elongated ones with two cups aligned to the slope, and wider ones with three cups. A testbed with the gripper attached to a robot facilitates fishcake gripping, classification, and automation testing. Fishcake recognition and gripping tests revealed challenges based on shape, width, and material. Despite difficulties, a commendable 100% success rate was achieved for the majority of fishcakes, showcasing the gripper’s effectiveness. Identified improvements include reducing the suction cup diameter and increasing pressure for enhanced gripping and classification in confined spaces. The study demonstrates the successful development of a gripper for versatile fishcake handling, particularly beneficial for small workshops. The identified improvements offer pathways to enhance efficiency in fishcake gripping and classification within limited spaces.

Ensemble of Deep Features for Breast Cancer Histopathological Image Classification

Abstract Analysis of histopathological images (HIs) is crucial for detecting breast cancer (BR). However, because they vary, it is still very difficult to extract well-designed elements. Deep learning (DL) is a recent development that is used to extract high-level features. However, DL techniques continue to confront several difficult problems, such as the need for sufficient training data for DL models, which reduces the classification findings. In this study, an ensemble deep transfer convolutional neural network is presented to address this problem. The pre-trained models (ResNet50 and MobileNet) are employed to extract high-level features by freezing the front layer parameters while fine-tuning the last layers. In the proposed ensemble framework, KNN, SVM, logistic regression and neural networks are used as base classifiers. The majority vote and product approaches are used to integrate the predictions of each separate classifier. In the benchmark BreaKHis dataset, the suggested ensemble model is compared to some current approaches. It demonstrates that while the ensemble model obtains a considerable accuracy of 97.72% for the multiclass classification test, it achieves an accuracy of 99.2% for the binary task. The suggested ensemble model’s effectiveness in extracting useful features for BR images is demonstrated by comparison with existing cutting-edge models.

Yogyakarta Batik Image Classification Based on Convolutional Neural Network

This paper studies the efficiency of identifying motifs and patterns in Yogyakarta batik using the Convolutional Neural Network (CNN) algorithm. This research uses the AlexNet architecture on CNN to increase the accuracy of batik image classification. Apart from that, it also involves the use of Canny edge detection techniques and feature extraction using the Gray Level Co-occurrence Matrix (GLCM) to improve the feature extraction process in batik images. There are 6 folders representing 6 types of motifs containing -+20 to 25 data that have been prepared for the training session. Next, the data is processed with 20% of the data used for training and 80% for testing. The accuracy of this research using the SGDM optimizer reached 100%. The evaluation results provide insight into the extent to which edge emphasis can improve the model's ability to recognize and classify batik patterns. It also presents classification test results and evaluation metrics such as precision, recall, and F1 score.

Explaining the Creativity Model of Students Based on the Perception of Competence with the Role of Perfectionism

Objective: Students endowed with creative and critical thinking, alongside their unique intellectual and behavioral styles, confront academic and social life challenges. They efficiently fulfill their duties by optimally utilizing their existential forces and spirituality. The purpose of this study was to investigate the development of a model of student creativity based on perceived competence with the mediation of perfectionism. Methods and Materials: The design of this study was descriptive-correlational. For this purpose, 300 students were selected using cluster random sampling and responded to the Abedi Creativity Scale (1993), the Harter's Self-Perception Profile for Adolescents (1982), and the Frost, Marten, Lahart, and Rosenblate Perfectionism Scale (1995). Data analysis was performed using path analysis. Additionally, SPSS 24 and AMOS 24 software were utilized for classification, processing, data analysis, and hypothesis testing. Findings: The evaluation of the hypothesized model using fit indices showed that the model fits the measurement model (CFI=0.93, NFI=0.91, RMSEA=0.07). Overall, in the proposed model of the study, the results indicated that perceived competence is related to creativity both directly and indirectly through perfectionism (p&lt;0.05). Moreover, the results showed a positive and significant relationship between perceived competence and perfectionism. Conclusion: Therefore, to enhance student creativity, intervening in perceived competence and perfectionism could be significant.

Exploration of the Integration of Traditional Music Cultural Elements in Music Informatization Teaching

Abstract With the development of information technology, music teaching methods are getting more affluent and more prosperous. This paper proposes a model for emotion classification and assessment that integrates traditional music culture elements with information technology in music teaching. The research first combines TextCNN and BiLSTM algorithms to establish the emotion classification model of conventional music. Then it combines PYIN and DTW algorithms to establish the evaluation model of traditional music, which completes the auxiliary efficacy of music informationized teaching. In the emotion classification test of the model, the classification accuracy and F1 value of emotions of different music samples are 82.98% and 75.22%, respectively. The model’s recognition accuracy of the four voices is 86.76%, and the overall effective scoring percentage is 81.98% under different playing abnormalities. This study has had an impact on traditional music evaluation. The model in this paper can be used to classify and evaluate emotions in conventional music, providing more intelligent and high-quality technical services for integrating traditional music into music teaching.

Artificial intelligence for a cardiac drug study

Machine learning that is the vital part of modern artificial intelligence was used to classify the potential effects of two drugs on induced pluripotent stem cell-derived cardiomyocytes (iPSC-CM). Peak data were detected from calcium transient signals of cycling iPSC-CMs first at baseline, second exposed to adrenaline and third exposed with one of the drugs either flecainide or carvedilol. Various machine learning classification tests were executed to evaluate whether a drug did or did not affect signal shape after exposing adrenaline or the drugs. According to our classification results, machine learning can be applied to analyze possible effects of drugs on induced pluripotent stem cell-derived cardiomyocytes derived from the catecholaminergic polymorphic ventricular tachycardia patients’ cell samples.

Molecular Gene Expression Testing to Identify Alzheimer's Disease with High Accuracy from Fingerstick Blood.

There is no molecular test for Alzheimer's disease (AD) using self-collected samples, nor is there a definitive molecular test for AD. We demonstrate an accurate and potentially definitive TempO-Seq® gene expression test for AD using fingerstick blood spotted and dried on filter paper, a sample that can be collected in any doctor's office or can be self-collected. Demonstrate the feasibility of developing an accurate test for the classification of persons with AD from a minimally invasive sample of fingerstick blood spotted on filter paper which can be obtained in any doctor's office or self-collected to address health disparities. Fingerstick blood samples from patients clinically diagnosed with AD, Parkinson's disease (PD), or asymptomatic controls were spotted onto filter paper in the doctor's office, dried, and shipped to BioSpyder for testing. Three independent patient cohorts were used for training/retraining and testing/retesting AD and PD classification algorithms. After initially identifying a 770 gene classification signature, a minimum set of 68 genes was identified providing classification test areas under the ROC curve of 0.9 for classifying patients as having AD, and 0.94 for classifying patients as having PD. These data demonstrate the potential to develop a screening and/or definitive, minimally invasive, molecular diagnostic test for AD and PD using dried fingerstick blood spot samples that are collected in a doctor's office or clinic, or self-collected, and thus, can address health disparities. Whether the test can classify patients with AD earlier then possible with cognitive testing remains to be determined.

Multivariate probabilistic assessment of a regional database in Copenhagen

Extensive research has been carried out to establish empirical equations between two geotechnical properties and in spite of the increase in correlation studies, a construction of a multivariate distribution function of more than two parameters is still rare. The objective of this study is to investigate the possibility of constructing two multivariate distribution functions, each consisting of 5 geotechnical parameters based on a) triaxial and soil classification tests and b) oedometer and soil classification tests in stiff over-consolidated clay till from Copenhagen. For this purpose, laboratory measurements from twenty-six sites in Copenhagen are utilized and two multivariate databases are constructed. The correlations among the ratio of deviatoric and mean effective stresses (q/p´), over-consolidation ratio (OCR), secant modulus at reference stress 100 kPa (E50ref), initial void ratio (eo) and liquid limit (LL) are investigated in multivariate model A, while multivariate model B demonstrates the correlations among oedometer modulus at 100 kPa reference stress (Eoed(ur)ref), over-consolidation ratio (OCR), permeability change index (Ck), initial void ratio (eo) and liquid limit (LL). The Nataf transformation model is used for the construction of the multivariate distributions, which are then used to simulate the correlations between geotechnical properties by generating artificial samples. Finally, the artificial samples are compared with the original database for an initial validation of the model. The constructed multivariate models obtained as a result of this study can act as prior for Bayesian updating in multivariate distribution functions when additional geotechnical tests are carried out.

Utilizing Generative Adversarial Network for Synthetic Image Generation to Address Imbalance Challenges in Chest X-Ray Image Classification

Deep learning-based classifiers need lots of image data to train. Unfortunately, not all real-world cases are supported by a huge amount of image data. One of the cases are images for classification of pneumonia infections with chest X-rays images. This study proposes a way of synthesizing chest X-rays with abnormal conditions in order to use the synthesized images for classification purposes. A GAN-based technique can generate synthetic images with greater quality that resemble original images thus can provide a more balanced data distribution than other approaches. To indirectly evaluate the quality of our GAN-based synthetic images, we used CNN-based classification architectures on diverse datasets. Three scenarios examined the effects of synthetic picture categorization. Scenario-1: adding 90% of synthesized images to the original images into the training dataset. Scenario-2: adding 50% of synthesized images to the original images. Scenario-3: adding 10% of synthesized image to the original images. The classification test revealed significantly increased F1 scores in all scenarios. Our study also emphasizes the significance of addressing the problem of imbalanced collections of chest X-ray images and the capability of GANs to alleviate this issue.

Are current regulatory log Kow cut-off values fit-for-purpose as a screening tool for bioaccumulation potential in aquatic organisms?

Persistent, Bioaccumulative and Toxic (PBT) and very Persistent and very Bioaccumulative (vPvB) are regulatory hazard categories that have been set to manage the possible risks to humans and the environment from these chemicals. In industrial chemicals regulations, their aquatic Bioaccumulation potential is usually assessed first with a screening based on the octanol/water partition coefficient (Kow). However, current log Kow cut-off values triggering classification, categorisation and/or further fish bioconcentration testing are not harmonised worldwide, and they have never been assessed for their regulatory relevance. In this study, the experimentally determined log Kow and fish bioconcentration factors (BCF) of 532 chemicals were compared. While the analysis underlined the robustness of using log Kow as a screening tool (5/532 were false negatives; log Kow: non-bioaccumulative, but BCF: bioaccumulative), it also demonstrated the conservatism of the cut-offs used worldwide. Indeed, many chemicals were deemed potentially Bioaccumulative based on log Kow when a fish bioaccumulation test showed no concern (false positives), therefore, leading to unnecessary use of vertebrate animals. Our analysis shows that the log Kow cut-off could be increased to 4.5 in all regions for all purposes without leading to a reduced protection of humans and the environment.

SIBI (Sistem Bahasa Isyarat Indonesia) berbasis Machine Learning dan Computer Vision untuk Membantu Komunikasi Tuna Rungu dan Tuna Wicara

The Indonesian Sign Language System (SIBI) is used to translate sign language into text or speech. SIBI helps improve communication between people using sign language and those who do not understand it. Unlike commonly used languages, SIBI sign language is less known to most people due to a lack of interest. To address this, an artificial intelligence-based application was developed, focusing on deep learning to recognize SIBI sign language hand movements in real-time. The model was created with 20 epochs, a batch size of 16, and a learning rate of 0.001. It consists of 13 layers, with the ReLU activation function used for the input layer, while the output layer uses Sigmoid. The ADAM optimizer was used to expedite the model creation process. The image dataset used had a size of 300x300 pixels. In the classification testing of the SIBI alphabet results in this study, it was tested using distance tests. The distance between the webcam and the SIBI language speaker was divided into two categories: 40 cm and 60 cm. For a 40-cm distance, an accuracy of 87.50% was obtained, and for a 60-cm distance, an accuracy of 79.17% was achieved. One limitation of this study is that two alphabets, J and Z, were not included in the dataset. This is because recognition of these two alphabets requires not only finger pattern recognition but also recognition of their gesture patterns.