Search results for: classification of matter

Authors: S. Nandagopalan, N. Pradeep

Abstract:

The aim of this paper is to propose a general framework for storing, analyzing, and extracting knowledge from two-dimensional echocardiographic images, color Doppler images, non-medical images, and general data sets. A number of high performance data mining algorithms have been used to carry out this task. Our framework encompasses four layers namely physical storage, object identification, knowledge discovery, user level. Techniques such as active contour model to identify the cardiac chambers, pixel classification to segment the color Doppler echo image, universal model for image retrieval, Bayesian method for classification, parallel algorithms for image segmentation, etc., were employed. Using the feature vector database that have been efficiently constructed, one can perform various data mining tasks like clustering, classification, etc. with efficient algorithms along with image mining given a query image. All these facilities are included in the framework that is supported by state-of-the-art user interface (UI). The algorithms were tested with actual patient data and Coral image database and the results show that their performance is better than the results reported already.

Keywords: active contour, Bayesian, echocardiographic image, feature vector

Procedia PDF Downloads 445

Authors: Obulesh Avuku, Satwik Sunnam, Sri Charan Mohan Janthuka, Keerthi Yalamaddi

Abstract:

In the United States alone, there are over 1.5 million spine fractures per year, resulting in about 17,730 spinal cord injuries. The cervical spine is where fractures in the spine most frequently occur. The prevalence of spinal fractures in the elderly has increased, and in this population, fractures may be harder to see on imaging because of coexisting degenerative illness and osteoporosis. Nowadays, computed tomography (CT) is almost completely used instead of radiography for the imaging diagnosis of adult spine fractures (x-rays). To stop neurologic degeneration and paralysis following trauma, it is vital to trace any vertebral fractures at the earliest. Many approaches have been proposed for the classification of the cervical spine [2d models]. We are here in this paper trying to break the bounds and use the vision transformers, a State-Of-The-Art- Model in image classification, by making minimal changes possible to the architecture of ViT and making it 3D-enabled architecture and this is evaluated using a weighted multi-label logarithmic loss. We have taken this problem statement from a previously held Kaggle competition, i.e., RSNA 2022 Cervical Spine Fracture Detection.

Keywords: cervical spine, spinal fractures, osteoporosis, computed tomography, 2d-models, ViT, multi-label logarithmic loss, Kaggle, public score, private score

Procedia PDF Downloads 114

Authors: Kazeem Adesina Dauda, Waheed Babatunde Yahya

Abstract:

To design a tree that maximizes within-node homogeneity, there is a need for a homogeneity measure that is appropriate for event history data with multiple risks. We consider the use of Deviance and Modified Cox-Snell residuals as a measure of impurity in Classification Regression Tree (CART) and compare our results with the results of Fiona (2008) in which homogeneity measures were based on Martingale Residual. Data structure approach was used to validate the performance of our proposed techniques via simulation and real life data. The results of univariate competing risk revealed that: using Deviance and Cox-Snell residuals as a response in within node homogeneity classification tree perform better than using other residuals irrespective of performance techniques. Bone marrow transplant data and double-blinded randomized clinical trial, conducted in other to compare two treatments for patients with prostate cancer were used to demonstrate the efficiency of our proposed method vis-à-vis the existing ones. Results from empirical studies of the bone marrow transplant data showed that the proposed model with Cox-Snell residual (Deviance=16.6498) performs better than both the Martingale residual (deviance=160.3592) and Deviance residual (Deviance=556.8822) in both event of interest and competing risks. Additionally, results from prostate cancer also reveal the performance of proposed model over the existing one in both causes, interestingly, Cox-Snell residual (MSE=0.01783563) outfit both the Martingale residual (MSE=0.1853148) and Deviance residual (MSE=0.8043366). Moreover, these results validate those obtained from the Monte-Carlo studies.

Keywords: within-node homogeneity, Martingale residual, modified Cox-Snell residual, classification and regression tree

Procedia PDF Downloads 272

Authors: M. N. Abu Seman, L. M. Kei, M. A. Yusoff

Abstract:

Forward Osmosis (FO) polyamide thin-film composite membranes have been prepared by inter facial polymerization using commercial UF polyethersulfone as membrane support. Different inter facial polymerization times (10s, 30s and 60s) in the organic solution containing trimesoyl chloride (TMC) at constant m-phenylenediamine (MPD) concentration (2% w/v) were studied. The synthesized polyamide membranes then tested for treatment of natural organic matter (NOM) and compared to commercial Cellulose TriAcetate (CTA) membrane. It was found that membrane prepared with higher reaction time (30 s and 60 s) exhibited better membrane performance (flux and humic acid removal) over commercial CTA membrane.

Keywords: cellulose triacetate, forward osmosis, humic acid, polyamide

Procedia PDF Downloads 491

Authors: Ji Wen Sun, Nan Ping Shen, Yi Bei Wu

Abstract:

This study used Nursing Interventions Classification (NIC-6th) to identify the interventions performed in a pediatric cardiology unit, and then to analysis its frequency, time and difficulty, so as to give a brief review on what our nurses have done. The research team selected a 35 beds pediatric cardiology unit, and drawn all the nursing interventions in the nursing record from our hospital information system (HIS) from 1 October 2015 to 30 November 2015, using NIC-6th to do the matching and then counting their frequencies. Then giving each intervention its own time and difficulty code according to NIC-6th. The results showed that nurses in pediatric cardiology unit performed totally 43 interventions from 5394 statements, and most of them were in RN(basic) education level needed and less than 15 minutes time needed. There still had some interventions just needed by a nursing assistant but done by nurses, which should call for nurse managers to think about the suitable staffing. Thus, counting the summary of the product of frequency, time and difficulty for each intervention of each nurse can know one's performance. Acknowledgement Clinical Management Optimization Project of Shanghai Shen Kang Hospital Development Center (SHDC2014615); Hundred-Talent Program of Construction of Nursing Plateau Discipline (hlgy16073qnhb).

Keywords: nursing interventions, nursing interventions classification, nursing record, pediatric cardiology

Procedia PDF Downloads 364

Authors: Abebe Mulugojam Negash, Yongbin Yu, Ekong Favour, Bekalu Nigus Dawit, Molla Woretaw Teshome, Aynalem Birtukan Yirga

Abstract:

Breast cancer remains a significant health concern, necessitating advancements in diagnostic methodologies. Addressing this, our paper confronts the notable challenges in breast cancer classification, particularly the imbalance in datasets and the constraints in the accuracy and interpretability of prevailing deep learning approaches. We proposed an attention-based residual neural network (ResNet), which effectively combines the robust features of ResNet with an advanced attention mechanism. Enhanced through strategic data augmentation and positive weight adjustments, this approach specifically targets the issue of data imbalance. The proposed model is tested on the BreakHis dataset and achieved accuracies of 99.00%, 99.04%, 98.67%, and 98.08% in different magnifications (40X, 100X, 200X, and 400X), respectively. We evaluated the performance by using different evaluation metrics such as precision, recall, and F1-Score and made comparisons with other state-of-the-art methods. Our experiments demonstrate that the proposed model outperforms existing approaches, achieving higher accuracy in breast cancer classification.

Keywords: residual neural network, attention mechanism, positive weight, data augmentation

Procedia PDF Downloads 101

Authors: Sultan Z. Mahmud, Emily C. Graff, Adil Bashir

Abstract:

Phosphoprotein enriched in astrocytes 15 kDa (PEA-15) is a multifunctional adapter protein which is associated with the regulation of apoptotic cell death. Recently it has been discovered that PEA-15 is crucial in normal neurodevelopment of domestic cats, a gyrencephalic animal model, although the exact function of PEA-15 in neurodevelopment is unknown. This study investigates how PEA-15 affects the blood-brain barrier (BBB) permeability in cat brain, which can cause abnormalities in tissue metabolite and energy supplies. Severe polymicrogyria and microcephaly have been observed in cats with a loss of function PEA-15 mutation, affecting the normal neurodevelopment of the cat. This suggests that the vital role of PEA-15 in neurodevelopment is associated with gyrification. Neurodevelopment is a highly energy demanding process. The mammalian brain depends on glucose as its main energy source. PEA-15 plays a very important role in glucose uptake and utilization by interacting with phospholipase D1 (PLD1). Mitochondria also plays a critical role in bioenergetics and essential to supply adequate energy needed for neurodevelopment. Cerebral blood flow regulates adequate metabolite supply and recent findings also showed that blood plasma contains mitochondria as well. So the BBB can play a very important role in regulating metabolite and energy supply in the brain. In this study the blood-brain permeability in cat brain was measured using MRI magnetization transfer (MT) effect on the perfusion signal. Perfusion is the tissue mass normalized supply of blood to the capillary bed. Perfusion also accommodates the supply of oxygen and other metabolites to the tissue. A fraction of the arterial blood can diffuse to the tissue, which depends on the BBB permeability. This fraction is known as water extraction fraction (EF). MT is a process of saturating the macromolecules, which has an effect on the blood that has been diffused into the tissue while having minimal effect on intravascular blood water that has not been exchanged with the tissue. Measurement of perfusion signal with and without MT enables to estimate the microvascular blood flow, EF and permeability surface area product (PS) in the brain. All the experiments were performed with Siemens 7T Magnetom with 32 channel head coil. Three control cats and three PEA-15 mutant cats were used for the study. Average EF in white and gray matter was 0.9±0.1 and 0.86±0.15 respectively, perfusion in white and gray matter was 85±15 mL/100g/min and 97±20 mL/100g/min respectively, PS in white and gray matter was 201±25 mL/100g/min and 225±35 mL/100g/min respectively for control cats. For PEA-15 mutant cats, average EF in white and gray matter was 0.81±0.15 and 0.77±0.2 respectively, perfusion in white and gray matter was 140±25 mL/100g/min and 165±18 mL/100g/min respectively, PS in white and gray matter was 240±30 mL/100g/min and 259±21 mL/100g/min respectively. This results show that BBB is compromised in PEA-15 mutant cat brain, where EF is decreased and perfusion as well as PS are increased in the mutant cats compared to the control cats. This findings might further explain the function of PEA-15 in neurodevelopment.

Keywords: BBB, cat brain, magnetization transfer, PEA-15

Procedia PDF Downloads 143

Authors: Nadjiba Mezıou-Cheboutı, Amel Merabet, Yahia Cheboutı, Nassima Behidj, Fatima-Zohra Bıssaad et Salahedine Doumandjı

Abstract:

Among the Anacardiaceae, the fruit (Pistacia vera L.) is the only species that produces edible fruits. The introduction of real pistachio was made in the early sixties by an FAO program in Algeria in several regions in the northern part of Algeria: Tlemcen, Sidi Bel Abbes, Batna, Bouira M'sila . Chemical analyzes of seeds pistachios were made on seeds from an orchard that localizes to Bechloul (Bouira) located in bioclimatic sub-humid temperate winter floor. Analyzes reveal dry matter content of 3.60 ± 0.45%, the water rate is 7.21 ± 0.36%. However, the fat content is 46.00 ± 0.90%, in average blood sugar, it is 4.02 ± 0.47%, the protein reached 29.88 ± 0.76%. Given the very interesting that high-fat food nutritional values, culture pistachio must be considered for its extension in Algeria.

Keywords: pistachio, dry matter, fat, sugar, protein

Procedia PDF Downloads 355

Authors: Bliss Singhal

Abstract:

Machine learning (ML) is a branch of Artificial Intelligence (AI) where computers analyze data and find patterns in the data. The study focuses on the detection of metastatic cancer using ML. Metastatic cancer is the stage where cancer has spread to other parts of the body and is the cause of approximately 90% of cancer-related deaths. Normally, pathologists spend hours each day to manually classifying whether tumors are benign or malignant. This tedious task contributes to mislabeling metastasis being over 60% of the time and emphasizes the importance of being aware of human error and other inefficiencies. ML is a good candidate to improve the correct identification of metastatic cancer, saving thousands of lives and can also improve the speed and efficiency of the process, thereby taking fewer resources and time. So far, the deep learning methodology of AI has been used in research to detect cancer. This study is a novel approach to determining the potential of using preprocessing algorithms combined with classification algorithms in detecting metastatic cancer. The study used two preprocessing algorithms: principal component analysis (PCA) and the genetic algorithm, to reduce the dimensionality of the dataset and then used three classification algorithms: logistic regression, decision tree classifier, and k-nearest neighbors to detect metastatic cancer in the pathology scans. The highest accuracy of 71.14% was produced by the ML pipeline comprising of PCA, the genetic algorithm, and the k-nearest neighbor algorithm, suggesting that preprocessing and classification algorithms have great potential for detecting metastatic cancer.

Keywords: breast cancer, principal component analysis, genetic algorithm, k-nearest neighbors, decision tree classifier, logistic regression

Procedia PDF Downloads 81

Authors: Zayneb Kthiri, Walid Hamada

Abstract:

The application of organic fertilizers is generally known to be useful to sustain soil fertility and plant growth, especially in poor soils, with less than 1% of organic matter, as it is very common in our Tunisian fields. Therefore, we focused on evaluating the effect of a new released liquid organic fertilizer named Solorga (with 5% of organic matter) compared to a reference product (Espartan: Kimitec, Spain) on tomato plant growth and physiology. Both fertilizers, derived from plant decomposition, were applied at an early stage in hydroponic system and under greenhouse. In hydroponic system, after 14 days of their application by root feeding, a significant difference was observed between treatments. Indeed, Solorga improved shoots and roots length, as well as the biomass respectively, by 45%, 27%, and 27.8% increase rate, while compared to control plants. However, Espartan induced less the measured parameters while compared to untreated control. Moreover, Solorga significantly increased the chlorophyll content by 42% compared to control and by 32% compared to Espartan. In the greenhouse, after 20 days of treatments, the results showed a significant effect of both fertilizers on SPAD index and the number of flowers blossom. Solorga increased the amount of chlorophyll present in the leaf by 7% compared to Espartan as well as the plant height under greenhouse. Moreover, the number of flowers blossom increased by 15% in plants treated with Solorga while compared to Espartan. Whereas, there is no notable difference between both organic fertilizers on the fruits blossom and the number of fruits per blossom. In conclusion, even though there is a difference in the organic matter between both fertilizers, Solorga improved better the plant growth in controlled conditions in hydroponic system while compared to Espartan. Altogether the obtained results are encouraging for the use of Solorga as a soil enriching source of organic matter to help plants to boost their growth and help them to overcome abiotic stresses linked to soil fertility.

Keywords: tomato, plant growth, organic fertilizer, hydroponic system, greenhouse

Procedia PDF Downloads 138

Authors: Sabir Hussain, Mujtaba Hassan, Kashif Rasool, Asif Shahzad

Abstract:

Fine particulate matter with an aerodynamic diameter of less than 2.5 μm (PM2.5) was collected in Islamabad from November 2022 to January 2023, at urban sites. The average mass concentrations of PM2.5 varied, ranging from 90.5 to 133 μg m−3 in urban areas. Environmental scanning electron microscopy (ESEM) analysis revealed that Islamabad's PM2.5 comprised soot aggregates, ashes, minerals, bio-particles, and unidentified particles. Results from inductively coupled plasma atomic emission spectroscopy (ICP-OES) indicated a gradual increase in total elemental concentrations in Islamabad PM2.5 in winter, with relatively high levels in December. Significantly different elemental compositions were observed in urban PM2.5. Enrichment factor (EF) analysis suggested that elements such as K, Na, Ca, Mg, Al, Fe, Ba, and Sr were of natural origin, while As, Cu, Zn, Pb, Cd, Mn, Ni, and Se originated from anthropogenic sources. Plasmid DNA assays demonstrated varying levels of potential toxicity in Islamabad PM2.5 collected from urban sites, as well as across different seasons. Notably, the urban winter PM2.5 sample exhibited much stronger toxicity compared to other samples. The presence of heavy metals in Islamabad PM2.5, including Cu, Zn, Pb, Cd, Cr, Mn, and Ni, may have synergistic effects on human health.

Keywords: islamabad particulate matter pm2.5, scanning electron microscopy with energy-dispersive x-ray spectroscopy(sem-eds), fourier transform infrared spectroscopy(ftir), inductively coupled plasma optical emission spectroscopy(icp-oes)

Procedia PDF Downloads 79

Authors: Mouli Banga, Chaya Ravindra

Abstract:

Wound classification is a crucial step in wound diagnosis. An effective classifier can aid wound specialists in identifying wound types with reduced financial and time investments, facilitating the determination of optimal treatment procedures. This study presents a deep neural network-based classifier that leverages wound images and their corresponding locations to categorize wounds into various classes, such as diabetic, pressure, surgical, and venous ulcers. By incorporating a developed body map, the process of tagging wound locations is significantly enhanced, providing healthcare specialists with a more efficient tool for wound analysis. We conducted a comparative analysis between two prominent convolutional neural network models, ResNet50 and MobileNetV2, utilizing a dataset of 730 images. Our findings reveal that the RestNet50 outperforms MovileNetV2, achieving an accuracy of approximately 90%, compared to MobileNetV2’s 83%. This disparity highlights the superior capability of ResNet50 in the context of this dataset. The results underscore the potential of integrating deep learning with spatial data to improve the precision and efficiency of wound diagnosis, ultimately contributing to better patient outcomes and reducing healthcare costs.

Keywords: wound classification, MobileNetV2, ResNet50, multimodel

Procedia PDF Downloads 32

Authors: Christian Neunzig, Simon Fahle, Jürgen Schulz, Matthias Möller, Bernd Kuhlenkötter

Abstract:

The widespread use of machine learning applications in production is significantly accelerated by improved computing power and increasing data availability. Predictive quality enables the assurance of product quality by using machine learning models as a basis for decisions on test results. The use of real Bosch production data based on geometric gauge blocks from machining, mating data from assembly and hydraulic measurement data from final testing of directional valves is a promising approach to classifying the quality characteristics of workpieces.

Keywords: machine learning, classification, predictive quality, hydraulics, supervised learning

Procedia PDF Downloads 213

Authors: Yuntao Liu, Lei Wang, Haoran Xia

Abstract:

Machine learning has shown extensive applications in the development of classification models for autism spectrum disorder (ASD) using neural image data. This paper proposes a fusion multi-modal classification network based on a graph neural network. First, the brain is segmented into 116 regions of interest using a medical segmentation template (AAL, Anatomical Automatic Labeling). The image features of sMRI and the signal features of fMRI are extracted, which build the node and edge embedding representations of the brain map. Then, we construct a dynamically updated brain map neural network and propose a method based on a dynamic brain map adjacency matrix update mechanism and learnable graph to further improve the accuracy of autism diagnosis and recognition results. Based on the Autism Brain Imaging Data Exchange I dataset(ABIDE I), we reached a prediction accuracy of 74% between ASD and TD subjects. Besides, to study the biomarkers that can help doctors analyze diseases and interpretability, we used the features by extracting the top five maximum and minimum ROI weights. This work provides a meaningful way for brain disorder identification.

Keywords: autism spectrum disorder, brain map, supervised machine learning, graph network, multimodal data, model interpretability

Procedia PDF Downloads 66

Authors: Weiwei Cui, Dejian Lin, Leigang Zhang, Yao Wang, Zheng Sun, Lianfeng Li

Abstract:

Built-in test (BIT) is an important technology in testability field, and it is widely used in state monitoring and fault diagnosis. With the improvement of modern equipment performance and complexity, the scope of BIT becomes larger, and it leads to the emergence of false alarm problem. The false alarm makes the health assessment unstable, and it reduces the effectiveness of BIT. The conventional false alarm suppression methods such as repeated test and majority voting cannot meet the requirement for a complicated system, and the intelligence algorithms such as artificial neural networks (ANN) are widely studied and used. However, false alarm has a very low frequency and small sample, yet a method based on ANN requires a large size of training sample. To recognize the false alarm, we propose a method based on multi-classification support vector machine (SVM) in this paper. Firstly, we divide the state of a system into three states: healthy, false-alarm, and faulty. Then we use multi-classification with '1 vs 1' policy to train and recognize the state of a system. Finally, an example of fault injection system is taken to verify the effectiveness of the proposed method by comparing ANN. The result shows that the method is reasonable and effective.

Keywords: false alarm, fault diagnosis, SVM, k-means, BIT

Procedia PDF Downloads 155

Authors: Zdenek Silar, Martin Dobrovolny

Abstract:

This article deals with the obstacle detection on a railway crossing (clearance detection). Detection is based on the optical flow estimation and classification of the flow vectors by K-means clustering algorithm. For classification of passing vehicles is used optical flow direction determination. The optical flow estimation is based on a modified Lucas-Kanade method.

Keywords: background estimation, direction of optical flow, K-means clustering, objects detection, railway crossing monitoring, velocity vectors

Procedia PDF Downloads 518

Authors: Nadjiba Meziou-Chebouti, Amel Merabet, Yahia Chebouti, Nassima Behidj, Salahedine Doumandji

Abstract:

Among the Anacardiaceae, the fruit (Pistacia vera L.) is the only species that produces edible fruits. The introduction of real pistachio was made in the early sixties by an FAO program in Algeria in several regions in the northern part of Algeria: Tlemcen, Sidi Bel Abbes, Batna, Bouira M'sila. Chemical analyzes of seeds pistachios were made on seeds from an orchard that localizes to Bechloul (Bouira) located in bioclimatic sub-humid temperate winter floor. Analyzes reveal dry matter content of 3.60±0.45%, the water rate is 7.21±0.36%. However, the fat content is 46.00±0.90%, in average blood sugar, it is 4.02±0.47%, the protein reached 29.88±0.76%. Given the very interesting that high-fat food nutritional values, culture pistachio must be considered for its extension in Algeria.

Keywords: pistachio, dry matter, fat, sugar, protein

Procedia PDF Downloads 367

Authors: H. Hotait, X. Chiementin, L. Rasolofondraibe

Abstract:

This paper presents a continuous work to detect the abnormal state in the rolling bearing by studying the vibration signature analysis and calculation of the remaining useful life. To achieve these aims, two methods; the first method is the classification to detect the degradation state by the AOM-OPTICS (Acousto-Optic Modulator) method. The second one is the prediction of the degradation state using least-squares support vector regression and then compared with the linear degradation model. An experimental investigation on ball-bearing was conducted to see the effectiveness of the used method by applying the acquired vibration signals. The proposed model for predicting the state of bearing gives us accurate results with the experimental and numerical data.

Keywords: bearings, automatization, optimization, prognosis, classification, defect detection

Procedia PDF Downloads 119

Authors: Daniel M. Wolff, Patricia Gray, Rafael de la Parra Venegas

Abstract:

An unsupervised machine listening system is constructed and applied to a dataset of 17,195 30-second marine hydrophone recordings. The system is then heuristically supplemented with anecdotal listening, contextual recording information, and supervised learning techniques to reduce the number of false positives. Features for classification are assembled by extracting the following data from each of the audio files: the spectral centroid, root-mean-squared values for each frequency band of a 10-octave filter bank, and mel-frequency cepstral coefficients in 5-second frames. In this way both time- and frequency-domain information are contained in the features to be passed to a clustering algorithm. Classification is performed using the k-means algorithm and then a k-nearest neighbors search. Different values of k are experimented with, in addition to different combinations of the available feature sets. Hypothesized class labels are 'primarily anthrophony' and 'primarily biophony', where the best class result conforming to the former label has 104 members after heuristic pruning. This demonstrates how a large audio dataset has been made more tractable with machine learning techniques, forming the foundation of a framework designed to acoustically monitor and gauge biological and anthropogenic activity in a marine environment.

Keywords: anthrophony, hydrophone, k-means, machine learning

Procedia PDF Downloads 170

Authors: S. Nandagopalan, N. Pradeep

Abstract:

The aim of this paper is to propose a general framework for storing, analyzing, and extracting knowledge from two-dimensional echocardiographic images, color Doppler images, non-medical images, and general data sets. A number of high performance data mining algorithms have been used to carry out this task. Our framework encompasses four layers namely physical storage, object identification, knowledge discovery, user level. Techniques such as active contour model to identify the cardiac chambers, pixel classification to segment the color Doppler echo image, universal model for image retrieval, Bayesian method for classification, parallel algorithms for image segmentation, etc., were employed. Using the feature vector database that have been efficiently constructed, one can perform various data mining tasks like clustering, classification, etc. with efficient algorithms along with image mining given a query image. All these facilities are included in the framework that is supported by state-of-the-art user interface (UI). The algorithms were tested with actual patient data and Coral image database and the results show that their performance is better than the results reported already.

Keywords: active contour, bayesian, echocardiographic image, feature vector

Procedia PDF Downloads 420

Authors: M. Abdullah, Al-Wadud

Abstract:

Sensor-based activity recognition systems usually accounts which sensors have been activated to perform an activity. The system then combines the conditional probabilities of those sensors to represent different activities and takes the decision based on that. However, the information about the sensors which are not activated may also be of great help in deciding which activity has been performed. This paper proposes an approach where the sensory data related to both usage and non-usage of objects are utilized to make the classification of activities. Experimental results also show the promising performance of the proposed method.

Keywords: Naïve Bayesian, based classification, activity recognition, sensor data, object-usage model

Procedia PDF Downloads 321

Authors: Christian Neunzig, Simon Fahle, Jürgen Schulz, Matthias Möller, Bernd Kuhlenkötter

Abstract:

Digitalisation in production technology is a driver for the application of machine learning methods. Through the application of predictive quality, the great potential for saving necessary quality control can be exploited through the data-based prediction of product quality and states. However, the serial use of machine learning applications is often prevented by various problems. Fluctuations occur in real production data sets, which are reflected in trends and systematic shifts over time. To counteract these problems, data preprocessing includes rule-based data cleaning, the application of dimensionality reduction techniques, and the identification of comparable data subsets to extract stable features. Successful process control of the target variables aims to centre the measured values around a mean and minimise variance. Competitive leaders claim to have mastered their processes. As a result, much of the real data has a relatively low variance. For the training of prediction models, the highest possible generalisability is required, which is at least made more difficult by this data availability. The implementation of a machine learning application can be interpreted as a production process. The CRoss Industry Standard Process for Data Mining (CRISP-DM) is a process model with six phases that describes the life cycle of data science. As in any process, the costs to eliminate errors increase significantly with each advancing process phase. For the quality prediction of hydraulic test steps of directional control valves, the question arises in the initial phase whether a regression or a classification is more suitable. In the context of this work, the initial phase of the CRISP-DM, the business understanding, is critically compared for the use case at Bosch Rexroth with regard to regression and classification. The use of cross-process production data along the value chain of hydraulic valves is a promising approach to predict the quality characteristics of workpieces. Suitable methods for leakage volume flow regression and classification for inspection decision are applied. Impressively, classification is clearly superior to regression and achieves promising accuracies.

Keywords: classification, CRISP-DM, machine learning, predictive quality, regression

Procedia PDF Downloads 144

Authors: Serge Provost, Yishan Zang

Abstract:

Following a review of various approaches that are utilized for classifying the tail behavior of a distribution, an easily implementable methodology that relies on an arctangent transformation is presented. The classification criterion is actually based on the difference between two specific quantiles of the transformed distribution. The resulting categories enable one to classify distributional tails as distinctly short, short, nearly medium, medium, extended medium and somewhat long, providing that at least two moments exist. Distributions possessing a single moment are said to be long tailed while those failing to have any finite moments are classified as having an extremely long tail. Several illustrative examples will be presented.

Keywords: arctangent transformation, tail classification, heavy-tailed distributions, distributional moments

Procedia PDF Downloads 120

Authors: Aishrith Rao

Abstract:

COVID 19 is a pandemic which has resulted in thousands of deaths around the world and a huge impact on the global economy. Testing is a huge issue as the test kits have limited availability and are expensive to manufacture. Using deep learning methods on radiology images in the detection of the coronavirus as these images contain information about the spread of the virus in the lungs is extremely economical and time-saving as it can be used in areas with a lack of testing facilities. This paper focuses on binary classification and multi-class classification of COVID 19 and other diseases such as pneumonia, tuberculosis, etc. Different deep learning methods such as VGG-19, COVID-Net, ResNET+ SVM, Deep CNN, DarkCovidnet, etc., have been used, and their accuracy has been compared using the Chest X-Ray dataset.

Keywords: deep learning, computer vision, radiology, COVID-19, ResNet, VGG-19, deep neural networks

Procedia PDF Downloads 160

Authors: Saba Ebrahimi, Hedieh Ashrafi

Abstract:

Predicting the cover type of forests is a challenge for natural resource managers. In this project, we aim to perform a comprehensive comparative study of two well-known classification methods, support vector machine (SVM) and decision tree (DT). The comparison is first performed among different types of each classifier, and then the best of each classifier will be compared by considering different evaluation metrics. The effect of boosting and bagging for decision trees is also explored. Furthermore, the effect of principal component analysis (PCA) and feature selection is also investigated. During the project, the forest cover-type dataset from the remote sensing and GIS program is used in all computations.

Keywords: classification methods, support vector machine, decision tree, forest cover-type dataset

Procedia PDF Downloads 217

Authors: Nazmul Shahadat, Anthony S. Maida

Abstract:

In recent years, hypercomplex inspired neural networks improved deep CNN architectures due to their ability to share weights across input channels and thus improve cohesiveness of representations within the layers. The work described herein studies the effect of replacing existing layers in an Axial Attention ResNet with their quaternion variants that use cross-channel weight sharing to assess the effect on image classification. We expect the quaternion enhancements to produce improved feature maps with more interlinked representations. We experiment with the stem of the network, the bottleneck layer, and the fully connected backend by replacing them with quaternion versions. These modifications lead to novel architectures which yield improved accuracy performance on the ImageNet300k classification dataset. Our baseline networks for comparison were the original real-valued ResNet, the original quaternion-valued ResNet, and the Axial Attention ResNet. Since improvement was observed regardless of which part of the network was modified, there is a promise that this technique may be generally useful in improving classification accuracy for a large class of networks.

Keywords: axial attention, representational networks, weight sharing, cross-channel correlations, quaternion-enhanced axial attention, deep networks

Procedia PDF Downloads 83

Authors: Amritpal Singh Nafria, Rohit Sharma, Md. Shami Ansari

Abstract:

Up to now only five different equations of motion can be derived from velocity time graph without needing to know the normal and frictional forces acting at the point of contact. In this paper we obtained all possible requisite conditions to be considering an equation as an equation of motion. After that we classified equations of motion by considering two equations as fundamental kinematical equations of motion and other three as additional kinematical equations of motion. After deriving these five equations of motion, we examine the easiest way of solving a wide variety of useful numerical problems. At the end of the paper, we discussed the importance and educational benefits of classification of equations of motion.

Keywords: velocity-time graph, fundamental equations, additional equations, requisite conditions, importance and educational benefits

Procedia PDF Downloads 787

Authors: Jerzy Banski

Abstract:

Small towns represent a key element of settlement structure and serve a number of important functions associated with the servicing of rural areas that surround them. It is in light of this that scientific studies have paid considerable attention to the functional structure of centers of this kind, as well as the relationships with both surrounding rural areas and other urban centers. But a preliminary to such research has typically involved attempts at classifying the urban centers themselves, with this also assisting with the planning and shaping of development policy on different spatial scales. The purpose of the work is to test out the methods underpinning three different classifications of small urban centers, as well as to offer a preliminary interpretation of the outcomes obtained. Research took in 722 settlement units in Poland, granted town rights and populated by fewer than 20,000 inhabitants. A morphologically-based classification making reference to the database of topographic objects as regards land cover within the administrative boundaries of towns and cities was carried out, and it proved possible to distinguish the categories of “housing-estate”, industrial and R&R towns, as well as towns characterized by dichotomy. Equally, a functional/morphological approach taken with the same database allowed for the identification – via an alternative method – of three main categories of small towns (i.e., the monofunctional, multifunctional or oligo functional), which could then be described in far greater detail. A third, multi-criterion classification made simultaneous reference to the conditioning of a structural, a location-related, and an administrative hierarchy-related nature, allowing for distinctions to be drawn between small towns in 9 different categories. The results obtained allow for multifaceted analysis and interpretation of the geographical differentiation characterizing the distribution of Poland’s urban centers across space in the country.

Keywords: small towns, classification, local planning, Poland

Procedia PDF Downloads 87

Authors: S. M. Ishtiaque, V. K. Yadav, S. D. Joshi, J. K. Chatterjee

Abstract:

The DIAMETRIC FAULTS system has been developed that captures a bi-directional image of yarn continuously in sequentially manner and provides the detailed classification of faults. A novel mathematical framework developed on the acquired bi-directional images forms the basis of fault classification in four broad categories, namely, Thick1, Thick2, Thin and Normal Yarn. A discretised version of Radon transformation has been used to convert the bi-directional images into one-dimensional signals. Images were divided into training and test sample sets. Karhunen–Loève Transformation (KLT) basis is computed for the signals from the images in training set for each fault class taking top six highest energy eigen vectors. The fault class of the test image is identified by taking the Euclidean distance of its signal from its projection on the KLT basis for each sample realization and fault class in the training set. Euclidean distance applied using various techniques is used for classifying an unknown fault class. An accuracy of about 90% is achieved in detecting the correct fault class using the various techniques. The four broad fault classes were further sub classified in four sub groups based on the user set boundary limits for fault length and fault volume. The fault cross-sectional area and the fault length defines the total volume of fault. A distinct distribution of faults is found in terms of their volume and physical dimensions which can be used for monitoring the yarn faults. It has been shown from the configurational based characterization and classification that the spun yarn faults arising out of mass variation, exhibit distinct characteristics in terms of their contours, sizes and shapes apart from their frequency of occurrences.

Keywords: Euclidean distance, fault classification, KLT, Radon Transform

Procedia PDF Downloads 265

Authors: Zsolt Kotroczó, Katalin Juhos, Áron Béni, Gábor Várbíró, Tamás Kocsis, István Fekete

Abstract:

Soil organic matter plays an extremely important role in the functioning and regulation processes of ecosystems. It follows that the C content of organic matter in soil is one of the most important indicators of soil fertility. Part of the carbon stored in them is returned to the atmosphere during soil respiration. Climate change and inappropriate land use can accelerate these processes. Our work aimed to determine how soil CO2 emissions change over ten years as a result of organic matter manipulation treatments. With the help of this, we were able to examine not only the effects of the different organic matter intake but also the effects of the different microclimates that occur as a result of the treatments. We carried out our investigations in the area of the Síkfőkút DIRT (Detritus Input and Removal Treatment) Project. The research area is located in the southern, hilly landscape of the Bükk Mountains, northeast of Eger (Hungary). GPS coordinates of the project: 47°55′34′′ N and 20°26′ 29′′ E, altitude 320-340 m. The soil of the area is Luvisols. The 27-hectare protected forest area is now under the supervision of the Bükki National Park. The experimental plots in Síkfőkút were established in 2000. We established six litter manipulation treatments each with three 7×7 m replicate plots established under complete canopy cover. There were two types of detritus addition treatments (Double Wood and Double Litter). In three treatments, detritus inputs were removed: No Litter No Roots plots, No Inputs, and the Controls. After the establishment of the plots, during the drier periods, the NR and NI treatments showed the highest CO2 emissions. In the first few years, the effect of this process was evident, because due to the lack of living vegetation, the amount of evapotranspiration on the NR and NI plots was much lower, and transpiration practically ceased on these plots. In the wetter periods, the NL and NI treatments showed the lowest soil respiration values, which were significantly lower compared to the Co, DW, and DL treatments. Due to the lower organic matter content and the lack of surface litter cover, the water storage capacity of these soils was significantly limited, therefore we measured the lowest average moisture content among the treatments after ten years. Soil respiration is significantly influenced by temperature values. Furthermore, the supply of nutrients to the soil microorganisms is also a determining factor, which in this case is influenced by the litter production dictated by the treatments. In the case of dry soils with a moisture content of less than 20% in the initial period, litter removal treatments showed a strong correlation with soil moisture (r=0.74). In very dry soils, a smaller increase in moisture does not cause a significant increase in soil respiration, while it does in a slightly higher moisture range. In wet soils, the temperature is the main regulating factor, above a certain moisture limit, water displaces soil air from the soil pores, which inhibits aerobic decomposition processes, and so heterotrophic soil respiration also declines.

Keywords: soil biology, organic matter, nutrition, DIRT, soil respiration

Procedia PDF Downloads 74