Classification Model Research Articles

Infrared spectroscopy as a new approach for early fabry disease screening: a pilot study

Abstract Background Fabry disease (FD) is a rare X-linked lysosomal storage disorder marked by alpha-galactosidase-A (α-Gal A) deficiency, caused by pathogenic mutations in the GLA gene, resulting in the accumulation of glycosphingolipids within lysosomes. The current screening test relies on measuring α-Gal A activity. However, this approach is limited to males. Infrared (IR) spectroscopy is a technique that can generate fingerprint spectra of a biofluid’s molecular composition and has been successfully applied to screen numerous diseases. Herein, we investigate the discriminating vibration profile of plasma chemical bonds in patients with FD using attenuated total reflection Fourier-transform IR (ATR-FTIR) spectroscopy. Results The Fabry disease group (n = 47) and the healthy control group (n = 52) recruited were age-matched (39.2 ± 16.9 and 36.7 ± 10.9 years, respectively), and females were predominant in both groups (59.6% and 65.4%, respectively). All patients had the classic phenotype (100%), and no late-onset phenotype was detected. A generated partial least squares discriminant analysis (PLS-DA) classification model, independent of gender, allowed differentiation of samples from FD vs. control groups, reaching 100% sensitivity, specificity and accuracy. Conclusion ATR-FTIR spectroscopy harnessed to pattern recognition algorithms can distinguish between FD patients and healthy control participants, offering the potential of a fast and inexpensive screening test.

Tree-based classification model for Long-COVID infection prediction with age stratification using data from the national COVID cohort collaborative (N3C)

Abstract Objectives We propose and validate a domain knowledge-driven classification model for diagnosing Post-acute sequelae of SARS-CoV-2 infection (PASC), also known as Long COVID, using Electronic Health Records (EHR) data. Materials and Methods We developed a robust model that incorporates features strongly indicative of PASC or associated with the severity of COVID-19 symptoms as identified in our literature review. The XGBoost tree-based architecture was chosen for its ability to handle class-imbalanced data and its potential for high interpretability. Using the training data provided by the Long COVID Computation Challenge (L3C), which was a sample of the National COVID Collaborative (N3C), our models were fine-tuned and calibrated to optimize Area Under the Receiver Operating Characteristic curve (AUROC) and the F1 score, following best practices for the class-imbalanced N3C data. Results Our age-stratified classification model demonstrated strong performance with an average 5-fold cross-validated AUROC of 0.844 and F1 score of 0.539 across the young adult, mid-aged, and older-aged populations in the training data. In an independent testing dataset, which was made available after the challenge was over, we achieved an overall AUROC score of 0.814 and F1 score of 0.545. Discussion The results demonstrated the utility of knowledge-driven feature engineering in a sparse EHR data and demographic stratification in model development to diagnose a complex and heterogeneously presenting condition like PASC. The model's architecture, mirroring natural clinician decision-making processes, contributed to its robustness and interpretability, which are crucial for clinical translatability. Further, the model’s generalizability was evaluated over a new cross-sectional data as provided in the later stages of the L3C challenge. Conclusion The study proposed and validated the effectiveness of age-stratified, tree-based classification models to diagnose PASC. Our approach highlights the potential of machine learning in addressing the diagnostic challenges posed by the heterogeneity of Long COVID symptoms.

Optimized biomedical entity relation extraction method with data augmentation and classification using GPT-4 and Gemini.

Despite numerous research efforts by teams participating in the BioCreative VIII Track 01 employing various techniques to achieve the high accuracy of biomedical relation tasks, the overall performance in this area still has substantial room for improvement. Large language models bring a new opportunity to improve the performance of existing techniques in natural language processing tasks. This paper presents our improved method for relation extraction, which involves integrating two renowned large language models: Gemini and GPT-4. Our new approach utilizes GPT-4 to generate augmented data for training, followed by an ensemble learning technique to combine the outputs of diverse models to create a more precise prediction. We then employ a method using Gemini responses as input to fine-tune the BioNLP-PubMed-Bert classification model, which leads to improved performance as measured by precision, recall, and F1 scores on the same test dataset used in the challenge evaluation. Database URL: https://biocreative.bioinformatics.udel.edu/tasks/biocreative-viii/track-1/.

A dual-labeled dataset and fusion model for automatic teeth segmentation, numbering, and state assessment on panoramic radiographs

BackgroundRecently, deep learning has been increasingly applied in the field of dentistry. The aim of this study is to develop a model for the automatic segmentation, numbering, and state assessment of teeth on panoramic radiographs.MethodsWe created a dual-labeled dataset on panoramic radiographs for training, incorporating both numbering and state labels. We then developed a fusion model that combines a YOLOv9-e instance segmentation model with an EfficientNetv2-l classification model. The instance segmentation model is used for tooth segmentation and numbering, whereas the classification model is used for state evaluation. The final prediction results integrate tooth position, numbering, and state information. The model’s output includes result visualization and automatic report generation.ResultsPrecision, Recall, mAP50 (mean Average Precision), and mAP50-95 for the tooth instance segmentation task are 0.989, 0.955, 0.975, and 0.840, respectively. Precision, Recall, Specificity, and F1 Score for the tooth classification task are 0.943, 0.933, 0.985, and 0.936, respectively.ConclusionsThis fusion model is the first to integrate automatic dental segmentation, numbering, and state assessment. It provides highly accurate results, including detailed visualizations and automated report generation.

VISTA: vision improvement via split and reconstruct deep neural network for fundus image quality assessment

AbstractWidespread eye conditions such as cataracts, diabetic retinopathy, and glaucoma impact people worldwide. Ophthalmology uses fundus photography for diagnosing these retinal disorders, but fundus images are prone to image quality challenges. Accurate diagnosis hinges on high-quality fundus images. Therefore, there is a need for image quality assessment methods to evaluate fundus images before diagnosis. Consequently, this paper introduces a deep learning model tailored for fundus images that supports large images. Our division method centres on preserving the original image’s high-resolution features while maintaining low computing and high accuracy. The proposed approach encompasses two fundamental components: an autoencoder model for input image reconstruction and image classification to classify the image quality based on the latent features extracted by the autoencoder, all performed at the original image size, without alteration, before reassembly for decoding networks. Through post hoc interpretability methods, we verified that our model focuses on key elements of fundus image quality. Additionally, an intrinsic interpretability module has been designed into the network that allows decomposing class scores into underlying concepts quality such as brightness or presence of anatomical structures. Experimental results in our model with EyeQ, a fundus image dataset with three categories (Good, Usable, and Rejected) demonstrate that our approach produces competitive outcomes compared to other deep learning-based methods with an overall accuracy of 0.9066, a precision of 0.8843, a recall of 0.8905, and an impressive F1-score of 0.8868. The code is publicly available at https://github.com/saifalkhaldiurv/VISTA_-Image-Quality-Assessment.

Coal-gangue sound recognition using hybrid multi-branch CNN based on attention mechanism fusion in noisy environments

The coal-gangue recognition technology plays an important role in the intelligent realization of fully mechanized caving face and the improvement of coal quality. Although great progress has been made for the coal-gangue recognition in recent years, most of them have not taken into account the impact of the complex environment of top coal caving on recognition performance. Herein, a hybrid multi-branch convolutional neural network (HMBCNN) is proposed for coal-gangue recognition, which based on improved Mel Frequency Cepstral Coefficient (MFCC) as well as Mel spectrogram, and attention mechanism. Firstly, the MFCC and its smooth feature matrix are input into each branch of one-dimensional multi-branch convolutional neural network, and the spliced features are extracted adaptively through multi-head attention mechanism. Secondly, the Mel spectrogram and its first-order derivative are input into each branch of the two-dimensional multi-branch convolutional neural network respectively, and the effective time-frequency information is paid attention to through the soft attention mechanism. Finally, at the decision-making level, the two networks are fused to establish a model for feature fusion and classification, obtaining optimal fusion strategies for different features and networks. A database of sound pressure signals under different signal-to-noise ratios and equipment operations is constructed based on a large amount of data collected in the laboratory and on-site. Comparative experiments and discussions are conducted on this database with advanced algorithms and different neural network structures. The results show that the proposed method achieves higher recognition accuracy and better robustness in noisy environments.

Association of Cannabis Use Reduction With Improved Functional Outcomes: An Exploratory Aggregated Analysis From Seven Cannabis Use Disorder Treatment Trials to Extract Data-Driven Cannabis Reduction Metrics.

This exploratory analysis sought to determine whether decreases in cannabis use are associated with improvements in cannabis-related problems and functional outcomes, and if so, what percentage decrease is associated with improvement. Data were aggregated from seven cannabis use disorder treatment trials conducted in the United States (N=920; ages 13 years and older; mean age, 25 years; 30% female, 7% Black, 11% Hispanic/Latinx). Outcome measures included the patient-reported Marijuana Problems Scale (MPS), Health-Related Quality of Life scale (HRQOL), and Pittsburgh Sleep Quality Index and the clinician-rated Clinical Global Impressions (CGI) severity and improvement scales (CGI-S and CGI-I). Generalized estimating equations tested the association between changes in 4-week cannabis use and improvements in functional outcomes. Classification and regression tree (CART) models were developed to determine what reductions in cannabis use could be used as classifiers of improvement. Decreases in the amount and frequency of cannabis use were significantly associated with improvements in MPS severity and total scores as well as improvements on the CGI-I and in sleep quality, but not improvements on the HRQOL. CART models performed best for CGI-I scores (72%-75% correct classification), while other outcome measures did not perform as well (40%-62% correct classification). CART models showed improvements on the CGI at 74% reduction in use amounts and 47% reduction in use days. Reductions in cannabis use (∼50% reduction in use days and ∼75% reduction in use amounts) were associated with clinician-assessed improvement, which suggests that cannabis use reduction may yield benefit among individuals with cannabis use disorder. These exploratory results extract a data-driven metric to inform future studies, clinicians, patients, and policy recommendations.

Vision transformer introduces a new vitality to the classification of renal pathology

Recent advancements in computer vision within the field of artificial intelligence (AI) have made significant inroads into the medical domain. However, the application of AI for classifying renal pathology remains challenging due to the subtle variations in multiple renal pathological classifications. Vision Transformers (ViT), an adaptation of the Transformer model for image recognition, have demonstrated superior capabilities in capturing global features and providing greater explainability. In our study, we developed a ViT model using a diverse set of stained renal histopathology images to evaluate its effectiveness in classifying renal pathology. A total of 1861 whole slide images (WSI) stained with HE, MASSON, PAS, and PASM were collected from 635 patients. Renal tissue images were then extracted, tiled, and categorized into 14 classes on the basis of renal pathology. We employed the classic ViT model from the Timm library, utilizing images sized 384 × 384 pixels with 16 × 16 pixel patches, to train the classification model. A comparative analysis was conducted to evaluate the performance of the ViT model against traditional convolutional neural network (CNN) models. The results indicated that the ViT model demonstrated superior recognition ability (accuracy: 0.96–0.99). Furthermore, we visualized the identification process of the ViT models to investigate potentially significant pathological ultrastructures. Our study demonstrated that ViT models outperformed CNN models in accurately classifying renal pathology. Additionally, ViT models are able to focus on specific, significant structures within renal histopathology, which could be crucial for identifying novel and meaningful pathological features in the diagnosis and treatment of renal disease.

MCM-Vb\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$V_b$$\\end{document}F: dance hand gestures recognition with vision based features

To digitize and preserve the cultural heritage in the form of Indian classical dance become apparent area of research. Sattriya classical dance of North-East India (Assam) is one of the eight Indian classical dance forms that requires immediate preservation. Sattriya classical dance consists of 29 Asamyukta hastas (single-hand gestures) and 14 Samyukta hastas (double-hand gestures). Moreover, the foundation of Samyukta hasta depends on understanding Asamyukta hasta. Therefore, the paper aims to classify single-hand gestures of Sattriya classical dance only. Although, a solution based on two level classification method to classify the Sattriya classical dance is available in recent literature, but it requires a trial and error method to select the optimized features. Since, Asamyukta hastas can appear closely similar to each other and therefore misclassification chances are very high. Thus, accuracy rate obtained for the two level classification method was only 75.45%. So, to address this issues in this paper, a Multilevel Classification Model with Vision based Features (MCM-Vb\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$V_b$$\\end{document}F) has been proposed to classify the Asamyukta hastas of Sattriya classical dance. This model uses two types of feature matching, viz., high-level feature matching and low-level feature matching. To extract the high-level features and low-level features different algorithm has been proposed. In this model, features are automatically selected. This proposed MCM-Vb\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$V_b$$\\end{document}F model is also tested on Asamyukta hasta mudras of Bharatanatyam classical dance of South India (Tamil Nadu). This model obtain an accuracy 94.12%, 87.14% for Sattriya classical dance Single-Hand Gestures (SSHG) dataset and Bharatnatyam classical dance Single-Hand Gestures (BHSG) dataset respectively. This paper also provides the comparative study of the proposed model MCM-Vb\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$V_b$$\\end{document}F with traditional bench-mark classifier model such as Naive Bayes, Decision Tree and Support Vector Classifier (SVM) etc.

Research on fiber OTDR signal recognition and classification based on the ViT model

In the field of event recognition for phase-sensitive optical time-domain reflectometry (-OTDR), convolutional neural networks (CNNs) have been the mainstream tool. However, Transformer models, with their self-attention mechanism, have provided a new perspective for image recognition tasks. This paper proposes a -OTDR event recognition method based on the Transformer model and experimentally demonstrates its significant advantages over CNNs. The method utilizes an attention-based feature extraction approach to recognize distributed fiber optic sensing signals in a phase-sensitive optical time-domain reflectometer (-OTDR). In perimeter defense applications, various interferences and noise severely affect the recognition of -OTDR sensing signals, leading to false alarms, making accurate signal recognition both challenging and necessary. Extracting signal features and using machine learning classification models to recognize signals has been a research hotspot in recent years. The effectiveness of feature extraction is crucial to classification performance. This project introduces a temporal-spatial dual attention mechanism into the feature extraction of fiber optic sensing signals to improve the accuracy and robustness of signal recognition.

Opportunities and limits of image-based plant stress phenotyping: detecting plant salt stress status using machine learning techniques

Abstract Plant stress phenotyping is a useful tool to facilitate targeted interventions and optimize management practices of plants growing in stressful conditions. However, various technical challenges still need to be overcome, though, and there is a clear need for investigating functional traits that can be used as a proxy for stress prediction, especially for abiotic stresses. This experimental work leveraged machine learning classification models to detect salt stress in two populations of a non-model species, combining image-based approaches (i.e. both manual and automated) and minimal morpho-physiological/biochemical analyses. A small set of specific features, combining malondialdehyde content with other non-destructive image-derived traits, such as Chroma Difference and Chroma Ratio indices, was able to distinguish non-stressed from stressed plants (2-class model; precision: 0.91) as well as stress intensity (3-class model; precision: 0.84). This flexible approach can be adapted to different plant developmental stages and leaf shape/morphology. In the future, the robustness and reliability of the models should be tested in other species and other abiotic stresses, such as drought.

Deciphering Cathepsin K inhibitors: a combined QSAR, docking and MD simulation based machine learning approaches for drug design.

Cathepsin K (CatK), a lysosomal cysteine protease, contributes to skeletal abnormalities, heart diseases, lung inflammation, and central nervous system and immune disorders. Currently, CatK inhibitors are associated with severe adverse effects, therefore limiting their clinical utility. This study focuses on exploring quantitative structure-activity relationships (QSAR) on a dataset of CatK inhibitors (1804) compiled from the ChEMBL database to predict the inhibitory activities. After data cleaning and pre-processing, a total of 1568 structures were selected for exploratory data analysis which revealed physicochemical properties, distributions and statistical significance between the two groups of inhibitors. PubChem fingerprinting with 11 different machine-learning classification models was computed. The comparative analysis showed the ET model performed well with accuracy values for the training set (0.999), cross-validation (0.970) and test set (0.977) in line with OECD guidelines. Moreover, to gain structural insights on the origin of CatK inhibition, 15 diverse molecules were selected for molecular docking. The CatK inhibitors (1 and 2) exhibited strong binding energies of -8.3 and -7.2 kcal/mol, respectively. MD simulation (300 ns) showed strong structural stability, flexibility and interactions in selected complexes. This synergy between QSAR, docking, MD simulation and machine learning models strengthen our evidence for developing novel and resilient CatK inhibitors.

Optimizing Plant Disease Classification with Hybrid Convolutional Neural Network–Recurrent Neural Network and Liquid Time-Constant Network

This paper addresses the practical challenge of detecting tomato plant diseases using a hybrid lightweight model that combines a Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN). Traditional image classification models demand substantial computational resources, limiting their practicality. This study aimed to develop a model that can be easily implemented on low-cost IoT devices while maintaining high accuracy with real-world images. The methodology leverages a CNN for extracting high-level image features and an RNN for capturing temporal relationships, thereby enhancing model performance. The proposed model incorporates a Closed-form Continuous-time Neural Network, a lightweight variant of liquid time-constant networks, and integrates Neural Circuit Policy to capture long-term dependencies in image patterns, reducing overfitting. Augmentation techniques such as random rotation and brightness adjustments were applied to the training data to improve generalization. The results demonstrate that the hybrid models outperform their single pre-trained CNN counterparts in both accuracy and computational cost, achieving a 97.15% accuracy on the test set with the proposed model, compared to around 94% for state-of-the-art pre-trained models. This study provides evidence of the effectiveness of hybrid CNN-RNN models in improving accuracy without increasing computational cost and highlights the potential of liquid neural networks in such applications.

Detection and classification on MRI images of brain tumor using YOLO NAS deep learning model

If a brain tumor is not properly diagnosed, it might result in fatal consequences and major health issues. As a result, a key component of diagnosis is the early identification of brain tumors and the precise categorization of brain tumor types. This research work focusses on detection and classification such as pituitary, meningioma, glioma, and non-tumorous tumors from brain tumor MRI image using YOLO NAS model. Advances in deep learning have led to an increasing awareness of computer-aided diagnosis technology. To improve classification performance, the input data set is acquired from the REMBRANDT repository using the Digital Database of Brain Tumor Magnetic Resonance Images. The primary phases of this task include segmentation, classification, and pre-processing. The researcher preprocessed the RGB image to exclude any undesired spots before locating the ROI. For each brain tumor image, we used the hybrid anisotropic diffusion filtering (HADF) technique to remove noise. The Encoder-Decoder Network (En–DeNet), which uses a deep neural network based on U-Net as the encoder and a pre-trained EfficientNet as the decoder, is then used to segment the MRI images. A proposed model was developed using a deep learning-based YOLO NAS technique to identify brain cancers from MRI images such as meningioma tumors, non-tumor, glioma tumors, and pituitary tumors. The suggested model's classification performance is weighed based on parameters such as precision, F1-score, sensitivity, accuracy, and specificity. The proposed fusion method YOLO NAS has improved classification results of accuracy (AC = 0.997%), specificity (SP = 0.985%), precision (PR = 0.982%), F1-score (F1 = 0.992%), and sensitivity (SE = 0.985%) using 2570 MRI data for training and 630 data for testing and validation of brain tumor cases. The results show that the brain tumor type classification system based on the proposed YOLO NAS technique works better than the DNN, PDCNN, DenseNet-161, and DCNN-SGD model classifiers.

Design of Chinese painting style classification model based on multi-layer aggregation CNN

This study delves deeply into exploring the artistic value of traditional Chinese painting (TCP) and aims to bridge the gap between its fundamental characteristics and the realm of human emotions. To achieve this, a novel convolutional neural network (CNN)-based classification model for TCP emotions is proposed. By thoroughly analyzing the distinct emotional mapping relationships inherent in TCP, a comprehensive framework is developed. Notably, emotional feature regions are accurately extracted using image saliency, and a multi-layer aggregation recalibrated emotional feature module is seamlessly integrated into the CNN network structure. This integration strengthens the activation of features that significantly influence TCP emotions. Moreover, a method of multi-category weighted activation localization is skillfully employed to classify Chinese painting emotions within the CMYK (cyan magenta yellow black) color space. The empirical results convincingly demonstrate that our algorithm surpasses existing approaches such as VGG-19, GoogLeNet, ResNet-50, and WSCNet in the task of TCP emotion recognition, achieving an impressive accuracy of 92.36%, the largest error is 0.191. This improvement signifies the advancements made by our model in accurately capturing and understanding the emotional nuances within TCP. By outperforming previous methods, our research contributes to the convergence of multimedia technology and cultural education.

Research on the Method of Imperfect Wheat Grain Recognition Utilizing Hyperspectral Imaging Technology

As the primary grain crop in China, wheat holds a significant position in the country’s agricultural production, circulation, consumption, and various other aspects. However, the presence of imperfect grains has greatly impacted wheat quality and, subsequently, food security. In order to detect perfect wheat grains and six types of imperfect grains, a method for the fast and non-destructive identification of imperfect wheat grains using hyperspectral images was proposed. The main contents and results are as follows: (1) We collected wheat grain hyperspectral data. Seven types of wheat grain samples, each containing 300 grains, were prepared to construct a hyperspectral imaging system for imperfect wheat grains, and visible near-infrared hyperspectral data from 2100 wheat grains were collected. The Savitzky–Golay algorithm was used to analyze the hyperspectral images of wheat grains, selecting 261 dimensional effective hyperspectral datapoints within the range of 420.61–980.43 nm. (2) The Successive Projections Algorithm was used to reduce the dimensions of the 261 dimensional hyperspectral datapoints, selecting 33 dimensional hyperspectral datapoints. Principal Component Analysis was used to extract the optimal spectral wavelengths, specifically selecting hyperspectral images at 647.57 nm, 591.78 nm, and 568.36 nm to establish the dataset. (3) Particle Swarm Optimization was used to optimize the Support Vector Machines model, Convolutional Neural Network model, and MobileNet V2 model, which were established to recognize seven types of wheat grains. The comprehensive recognition rates were 93.71%, 95.14%, and 97.71%, respectively. The results indicate that a larger model with more parameters may not necessarily yield better performance. The research shows that the MobileNet V2 network model exhibits superior recognition efficiency, and the integration of hyperspectral image technology with the classification model can accurately identify imperfect wheat grains.

Intraoperative rapid assessment of the deep muscle surgical margin of tongue squamous cell carcinoma via Raman spectroscopy

PurposeAn accurate assessment of the surgical margins of tongue squamous cell carcinoma (TSCC), especially the deep muscle tissue, can help completely remove the cancer cells and thus minimize the risk of recurrence. This study aimed to develop a classification model that classifies TSCC and normal tissues in order to aid in the rapid and accurate intraoperative assessment of TSCC surgical deep muscle tissue margins.Materials and methodsThe study obtained 240 Raman spectra from 60 sections (30 TSCC and 30 normal) from 15 patients diagnosed with TSCC. The classification model based on the analysis of Raman spectral data was developed, utilizing principal component analysis (PCA) and linear discriminant analysis (LDA) for the diagnosis and classification of TSCC. The leave-one-out cross-validation was employed to estimate and evaluate the prediction performance model.ResultsThis approach effectively classified TSCC tissue and normal muscle tissue, achieving an accuracy of exceeding 90%. The Raman analysis showed that TSCC tissues contained significantly higher levels of proteins, lipids, and nucleic acids compared to the adjacent normal tissues. In addition, we have also explored the potential of Raman spectroscopy in classifying different histological grades of TSCC.ConclusionThe PCA–LDA tissue classification model based on Raman spectroscopy exhibited good accuracy, which could aid in identifying tumor-free margins during surgical interventions and present a promising avenue for the development of rapid and accurate intraoperative techniques.

Preprocessing Text Data for the Doc2Vec Model in the Automatic Classification of Tickets in Support Centers of Software Providers

Leading software providers typically implement customer technical support functions, which are crucial for promoting and enhancing the competitiveness of their products and services in global markets. The high volume and heterogeneity of support tickets (functional, temporal, linguistic, etc.) highlight the importance of efficient classification systems. Effective classification optimizes the distribution of these tickets among support center specialists and automates their processing using an established knowledge base. However, classifying these tickets is a loosely formalized task. For companies that have accumulated substantial data on customer requests, automating classification through machine learning methods and natural language processing models, such as Word2Vec, FastText, BERT, and GPT, becomes feasible. It is generally accepted that classification effectiveness primarily depends on the model employed. Nevertheless, the quality of these models is significantly influenced by the nature of the training data. Literature review of the reveals significant research interest in methods for the automatic classification of tickets specifically tailored to the operational conditions of software provider support centers. However, there is a noticeable gap in the literature regarding the impact of data preprocessing on the quality of these models. The article aims to clarify the techniques of data preprocessing and analyze their impact on the effectiveness of text classification, considering the specificity of software provider support centers. This study examines the stages of the automatic classification process for tickets, accounting for the unique characteristics of the data (customer text requests). A relevant set of specified methodological and instrumental tools was developed and tested using open data from a global software provider (DevExpress). The testing involved a database of 165,000 tickets. The study's results indicate that preprocessing can improve classification metrics such as F-measure, Precision, and Recall from 77% to 79%. Additionally, preprocessing significantly reduces the dimensionality of text data (by 48.2%) and increases model training speed (by 26.5%) without loss of accuracy, ensuring cost-efficiency and operational efficiency in the use of computational resources.

IHBOT: An Intelligent and Hybrid Model for Investigation and Classification of IoT Botnet

The Internet of Things (IoT) is revolutionizing the technological market with exponential growth year wise. This revolution of IoT applications has also brought hackers and malware to gain remote access to IoT devices. The security of IoT systems has become more critical for consumers and businesses because of their inherent heterogenous design and open interfaces. Since the release of Mirai in 2016, IoT malware has gained an exponential growth rate. As IoT system and their infrastructure have become critical resources that triggers IoT malware injected by various shareholders in different settings. The enormous applications cause flooding of insecure packets and commands that fueled threats for IoT applications. IoT botnet is one of the most critical malwares that keeps evolving with the network traffic and may harm the privacy of IoT devices. In this work, we presented several sets of malware analysis mechanisms to understand the behavior of IoT malware. We devise an intelligent and hybrid model (IHBOT) that integrates the malware analysis and distinct machine learning algorithms for the identification and classification of the different IoT malware family based on network traffic. The clustering mechanism is also integrated with the proposed model for the identification of malware families based on similarity index. We have also applied YARA rules for the mitigation of IoT botnet traffic.

Impact of the fibrosis-4 index in patients with ST-elevated myocardial infarction.

The fibrosis-4 (FIB4) index, a simple, noninvasive marker used for hepatic diseases, represents adverse outcomes. The aim of the present study was to evaluate whether the FIB4 index can predict adverse outcomes in patients with ST-elevation myocardial infarction (STEMI). We investigated patients with STEMI who underwent primary percutaneous coronary intervention (PCI) and were alive at discharge. The cut-off FIB4 index at discharge was investigated using the survival classification and regression tree (CART) model to predict adverse outcomes. The primary outcome was all-cause mortality. Between January 2006 and December 2018, 1354 patients with STEMI (median age, 68 years; men, 76.1%) were investigated. The median value of the FIB4 index was 1.21 (0.84-1.78). The CART model divided the study population into low (FIB4 index <0.945; n = 435), intermediate (0.945 ≤ FIB4 index < 2.185; n = 692), and high (FIB4 index ≥2.185; n = 227) groups based on the significant predictive values for all-cause death. During a median follow-up period of 4.3 years, all-cause death occurred in 208 patients (15.4%). The Kaplan-Meier analysis showed a significant increase in mortality with higher FIB4 index values (log-rank, P < 0.001). The multivariate Cox regression model revealed that the FIB4 index was an independent risk predictor for all-cause death in patients with STEMI [low group as reference vs. intermediate group, hazard ratio: 1.975; 95% confidence interval (CI): 1.166-3.346; P = 0.011 and vs. high group, hazard ratio: 4.633; 95% CI: 2.549-8.418; P < 0.001]. The FIB4 index was associated with the risk of all-cause mortality in patients with STEMI who underwent primary PCI.