Classification Accuracy Of Data Research Articles

An Efficient and Versatile Variational Method for High-Dimensional Data Classification

High-dimensional data classification is a fundamental task in machine learning and imaging science. In this paper, we propose an efficient and versatile multi-class semi-supervised classification method for classifying high-dimensional data and unstructured point clouds. To begin with, a warm initialization is generated by using a fuzzy classification method such as the standard support vector machine or random labeling. Then an unconstraint convex variational model is proposed to purify and smooth the initialization, followed by a step which is to project the smoothed partition obtained previously to a binary partition. These steps can be repeated, with the latest result as a new initialization, to keep improving the classification quality. We show that the convex model of the smoothing step has a unique solution and can be solved by a specifically designed primal–dual algorithm whose convergence is guaranteed. We test our method and compare it with the state-of-the-art methods on several benchmark data sets. Thorough experimental results demonstrate that our method is superior in both the classification accuracy and computation speed for high-dimensional data and point clouds.

A Refined Approach Involving Feature Correlation Analysis Followed by Classification using Python for Gene Assortment

Objectives: To make a study on hybrid scheme of selecting a most representative subset of genes to improve microarray data classification accuracy using machine learning algorithm. Method: ML-based SGSA approach proposed here has two parts of execution on gene expression datasets. First it utilizes an entropy based IGKullback–Leibler divergence to select the most informative genes. Subsequently, an attribute evaluation performed using correlation-based feature selection. After that a random forest based classifier is employed with 10-fold cross-validation. The proposed method involves data pre processing, testing, training, fitting an algorithm and then finds the best accuracy with comparable CPU cost. Findings: The rationale behind this study is that only most informative genes are submitted to classifier for classification task. Proven accuracy by this approach is 98.48 for Lymphoma, 89.69 for Breast, 86.67 for CNS, 97.22 for Leukemia, 98.4 for Lung cancer, 96.6 for MLL, 97.21 for Ovarian cancer and 98.5 for SRBCT over traditional machine learning algorithms like naïve bias, J48 and SMO. These results demonstrate the effectiveness of the suggested approach in accurately classifying tumors. The numerical illustration also showed that the new estimator is more efficient in terms of CPU cost. Novelty: The major problem with traditional machine learning algorithms is that all features are treated equally important during the classification process which makes them susceptible to the influence of outliers and difficult to find a meaningful class in the dataset. In this study, classification accuracy is improved by processing the most informative features in the classifiers. The primary contribution of the research is a hybrid ML model which uses IG based feature selection followed by correlation analysis. The result is then fed to RF based classifier which significantly enhance accuracy as well as CPU cost. Keywords: Machine Learning, Classification, Correlation, Feature Selection, Gene Expression

Multi-Feature Cross Attention-Induced Transformer Network for Hyperspectral and LiDAR Data Classification

Transformers have shown remarkable success in modeling sequential data and capturing intricate patterns over long distances. Their self-attention mechanism allows for efficient parallel processing and scalability, making them well-suited for the high-dimensional data in hyperspectral and LiDAR imagery. However, further research is needed on how to more deeply integrate the features of two modalities in attention mechanisms. In this paper, we propose a novel Multi-Feature Cross Attention-Induced Transformer Network (MCAITN) designed to enhance the classification accuracy of hyperspectral and LiDAR data. The MCAITN integrates the strengths of both data modalities by leveraging a cross-attention mechanism that effectively captures the complementary information between hyperspectral and LiDAR features. By utilizing a transformer-based architecture, the network is capable of learning complex spatial-spectral relationships and long-range dependencies. The cross-attention module facilitates the fusion of multi-source data, improving the network’s ability to discriminate between different land cover types. Extensive experiments conducted on benchmark datasets demonstrate that the MCAITN outperforms state-of-the-art methods in terms of classification accuracy and robustness.

Comprehensive application of AI algorithms with TCR NGS data for glioma diagnosis

T-cell receptor (TCR) detection can examine the extent of T-cell immune responses. Therefore, the article analyzed characteristic data of glioma obtained by DNA-based TCR high-throughput sequencing, to predict the disease with fewer biomarkers and higher accuracy. We downloaded data online and obtained six TCR-related diversity indices to establish a multidimensional classification system. By comparing actual presence of the 602 correlated sequences, we obtained two-dimensional and multidimensional datasets. Multiple classification methods were utilized for both datasets with the classification accuracy of multidimensional data slightly less to two-dimensional datasets. This study reduced the TCR β sequences through feature selection methods like RFECV (Recursive Feature Elimination with Cross-Validation). Consequently, using only the presence of these three sequences, the classification AUC value of 96.67% can be achieved. The combination of the three correlated TCR clones obtained at a source data threshold of 0.1 is: CASSLGGNTEAFF_TRBV12_TRBJ1-1, CASSYSDTGELFF_TRBV6_TRBJ2-2, and CASSLTGNTEAFF_TRBV12_TRBJ1-1. At 0.001, the combination is: CASSLGETQYF_TRBV12_TRBJ2-5, CASSLGGNQPQHF_TRBV12_TRBJ1-5, and CASSLSGNTIYF_TRBV12_TRBJ1-3. This method can serve as a potential diagnostic and therapeutic tool, facilitating diagnosis and treatment of glioma and other cancers.

Identification and Removal of Pollen Spectral Interference in the Classification of Hazardous Substances Based on Excitation Emission Matrix Fluorescence Spectroscopy.

Sensitively detecting hazardous and suspected bioaerosols is crucial for safeguarding public health. The potential impact of pollen on identifying bacterial species through fluorescence spectra should not be overlooked. Before the analysis, the spectrum underwent preprocessing steps, including normalization, multivariate scattering correction, and Savitzky-Golay smoothing. Additionally, the spectrum was transformed using difference, standard normal variable, and fast Fourier transform techniques. A random forest algorithm was employed for the classification and identification of 31 different types of samples. The fast Fourier transform improved the classification accuracy of the sample excitation-emission matrix fluorescence spectrum data by 9.2%, resulting in an accuracy of 89.24%. The harmful substances, including Staphylococcus aureus, ricin, beta-bungarotoxin, and Staphylococcal enterotoxin B, were clearly distinguished. The spectral data transformation and classification algorithm effectively eliminated the interference of pollen on other components. Furthermore, a classification and recognition model based on spectral feature transformation was established, demonstrating excellent application potential in detecting hazardous substances and protecting public health. This study provided a solid foundation for the application of rapid detection methods for harmful bioaerosols.

Evidential dissonance measure in robust multi-view classification to resist adversarial attack

Multi-view learning is effective in improving data classification accuracy through integrating information from multiple sources. To guarantee the reliability of multi-view classification, the trusted multi-view classification methods have been explored. However, extant trusted multi-view classification methods are still vulnerable to the low-quality views containing adversarial samples. This vulnerability arises from a challenge in accurately assessing the quality of data views that are subjected to adversarial attacks. To address this issue, we propose a robust multi-view classification method with dissonance measure of adversarial samples. Specifically, the proposed method utilizes the evidential dissonance measure in subjective logic to assess the quality of data views when encountering adversarial attacks. Based on the dissonance measure, we further propose a strategy of dissonance-aware belief integration for multi-view information fusion and construct the inter-view evidential gradient penalty in the learning objective of multi-view classification to improve the model robustness to resist adversarial samples. Experiments on diverse multi-view datasets confirm the reliability and robustness of the proposed multi-view classification method.

Artificial intelligence based social robots in the process of student mental health diagnosis

This paper in order to achieve the application of artificial intelligence based social robots in the process of student mental health diagnosis. When designing the architecture of social robots, factors such as interactivity, adaptability, and scalability were taken into consideration to ensure that they possess human like interaction characteristics and flexibility. Subsequently, a model was constructed based on deep learning technology to achieve functions such as sentiment classification, text mining, and optimization strategies. The input data set of the training model comes from the user’s interaction records and behavior data on the Internet social platform, as well as the user’s feedback information in the process of using the robot. The research on psychological data classification has constructed corresponding algorithms based on pointer networks and text models to achieve text feature extraction and classification. The psychological emotion mining module extracts emotional states from user discourse and maps them to corresponding categories of psychological problems. Finally, based on the user input question content, classify and optimize psychological problems. Research has shown that the robot has certain accuracy and practicality in data classification and student mental health diagnosis.

Effective Training and Inference Strategies for Point Classification in LiDAR Scenes

Light Detection and Ranging systems serve as robust tools for creating three-dimensional representations of the Earth’s surface. These representations are known as point clouds. Point cloud scene segmentation is essential in a range of applications aimed at understanding the environment, such as infrastructure planning and monitoring. However, automating this process can result in notable challenges due to variable point density across scenes, ambiguous object shapes, and substantial class imbalances. Consequently, manual intervention remains prevalent in point classification, allowing researchers to address these complexities. In this work, we study the elements contributing to the automatic semantic segmentation process with deep learning, conducting empirical evaluations on a self-captured dataset by a hybrid airborne laser scanning sensor combined with two nadir cameras in RGB and near-infrared over a 247 km2 terrain characterized by hilly topography, urban areas, and dense forest cover. Our findings emphasize the importance of employing appropriate training and inference strategies to achieve accurate classification of data points across all categories. The proposed methodology not only facilitates the segmentation of varying size point clouds but also yields a significant performance improvement compared to preceding methodologies, achieving a mIoU of 94.24% on our self-captured dataset.

Cost-effective data classification storage through text seasonal features

Data classification storage has emerged as an effective strategy, harnessing the diverse performance attributes of storage devices and orchestrating a harmonious equilibrium between energy consumption, cost considerations, and user accessibility. As research on emerging storage media (e.g. Non-Volatile Memory (NVM)) delves deeper, in scenarios characterized by dynamically evolving storage demands, conventional paradigms of rigid classification strategy and Solid-State Drive (SSD) and Hard Disk Drive (HDD) storage architectures fall short of addressing such complex situations. In this paper, we propose an effective data classification storage method using text seasonal features based on the traditional access frequency analysis strategy. First, to procure richer semantic information, we employ external knowledge actualizing the short-text feature expansion. Then, we leverage the ensemble learning stacking method optimized models to improve the accuracy of data classification based on seasonal features. Additionally, following the seasonal feature, we further classify the data into hot, warm, and cold and place them separately on the NVM, SSD, and HDD, which conserves storage energy consumption and operational costs while ensuring the quality of user access. The experimental results demonstrate that data classification accuracy can reach more than 95.10%, and the energy consumption and operating cost can be reduced by more than 30.22% and 8.73%, respectively.

An improve unsupervised discretization using optimization algorithms for classification problems

This paper addresses the classification problem in machine learning, focusing on predicting class labels for datasets with continuous features. Recognizing the critical role of discretization in enhancing classification performance, the study integrates equal width binning (EWB) with two optimization algorithms: the bat algorithm (BA), referred to as EB, and the whale optimization algorithm (WOA), denoted as EW. The primary objective is to determine the optimal technique for predicting relevant class labels. The paper emphasizes the significance of discretization in data preprocessing, offering a comprehensive approach that combines discretization techniques with optimization algorithms. An investigative study was undertaken to assess the efficiency of EB and EW by evaluating their classification performance using Naive Bayes and K-nearest neighbor algorithms on four continuous datasets sourced from the UCI datasets. According to the experimental findings, the suggested EB has a major effect on the accuracy, recall, and F-measure of data classification. The classification performance using EB outperforms other existing approaches for all datasets.

SIMILARITY-BASED ROUGH SET APPROACH IN INCOMPLETE INFORMATION SYSTEM USING POSSIBLE EQUIVALENT VALUE-SET

Data analytics generally helps businesses or entities to make better and efficient decision making. But in the face of growing volume of data or information, it becomes challenging to achieve these goals. One of which is on classification of information with high accuracy. Furthermore, when the information is incomplete, definitely it is more challenging in order to classify the information with high accuracy. Although incomplete information is well discussed using rough set theory for data classification, such as based on tolerance and similarity relations, there are still issues on accuracy to evaluate data classification. The main objective of this paper is to introduce a new similarity approach with semantically justified based on possible equivalent value-set related to incomplete information systems. It is based on a classification of three semantics types of incomplete information i.e., “any value”, “maybe value” and “not applicable value” for modelling similarity. Subsequently, the similarity precision between objects in incomplete information systems is considered. The comparative studies and simulation results between the proposed approach in terms of accuracy on synthetic data, four well-known classification datasets and one real marine dataset are presented. The proposed approach improves the accuracy up to two orders of magnitude and, thus verifying its data classification accuracy.

Enhancement of the Classification Performance of Fuzzy C-Means through Uncertainty Reduction with Cloud Model Interpolation

As an information granulation technology, clustering plays a pivotal role in unsupervised learning, serving as a fundamental cornerstone for various data mining techniques. The effective and accurate classification of data is a central focus for numerous researchers. For a dataset, we assert that the classification performance of a clustering method is significantly influenced by uncertain data, particularly those situated at the cluster boundaries. It is evident that uncertain data encapsulate richer information compared with others. Generally, the greater the uncertainty, the more information the data holds. Therefore, conducting a comprehensive analysis of this particular subset of data carries substantial significance. This study presents an approach to characterize data distribution properties using fuzzy clustering and defines the boundary and non-boundary characteristics (certainty and uncertainty) of the data. To improve the classification performance, the strategy focuses on reducing the uncertainty associated with boundary data. The proposed scheme involves inserting data points with the cloud computing technology based on the distribution characteristics of the membership functions to diminish the uncertainty of uncertain data. Building upon this, the contribution of boundary data is reassigned to the prototype in order to diminish the proportion of uncertain data. Subsequently, the classifier is optimized through data label (classification error) supervision. Ultimately, the objective is to leverage clustering algorithms for classification, thereby enhancing overall classification accuracy. Experimental results substantiate the effectiveness of the proposed scheme.

Research on improving accuracy and efficiency of animal data collection and classification using machine learning

With the continuous expansion of machine learning algorithms in various application domains, the application value of new algorithms, such as Support Vector Machines and Convolutional Neural Networks in data classification, has garnered increasing attention. This paper takes machine learning algorithms as the research entry point, explores the concept of machine learning, and delves into its application value in data classification. This paper, starting with an overview of machine learning algorithms, analyzes the supervised and unsupervised learning problems in machine learning, focusing on the applications of Convolutional Neural Networks, Support Vector Machine models, and logistic regression algorithms in data classification. This study emphasizes designing and implementing a machine learning-based image classification system. Through an in-depth exploration of the application of machine learning algorithms in data classification, a fully functional system is constructed, encompassing multiple modules, including machine vision and software development. This system accurately classifies and recognizes images, providing practical tools and technical support for image processing and analysis. In this study, the goal of achieving good image classification is realized through research and the application of machine learning algorithms. By designing and implementing a machine learning-based image classification system, the accuracy and efficiency of classification in handling massive data are improved. This system also demonstrates wide-ranging prospects in software development and machine vision, among other fields.

Investigation and implementation of digital software architecture based on internet of things

Application systems are becoming increasingly complex, and software system architecture plays an increasingly important role in the software development process. In view of the problems of the traditional software architecture, such as slow data processing speed, poor stability of the architecture and low accuracy of data classification, this paper studied the digital software architecture based on the Internet of Things (IoT) technology, and conducted data pre-processing on the data collected by sensors, mainly including data reading, data cleaning, data transformation and data reduction. Through experiments, it could be found that the accuracy of digital software architecture using data preprocessing for data classification was more than 93.52%, and the average accuracy of 10 experiments was 94.55%. At the same time, the digital software architecture based on the IoT was tested, and it was found that both the execution time and the original memory were better than other digital software architecture. The overall performance of digital software architecture based on the IoT was better.

Evaluation of the learning state of online video courses based on functional near infrared spectroscopy.

Studying brain activity during online learning will help to improve research on brain function based on real online learning situations, and will also promote the scientific evaluation of online education. Existing research focuses on enhancing learning effects and evaluating the learning process associated with online learning from an attentional perspective. We aimed to comparatively analyze the differences in prefrontal cortex (PFC) activity during resting, studying, and question-answering states in online learning and to establish a classification model of the learning state that would be useful for the evaluation of online learning. Nineteen university students performed experiments using functional near-infrared spectroscopy (fNIRS) to monitor the prefrontal lobes. The resting time at the start of the experiment was the resting state, watching 13 videos was the learning state, and answering questions after the video was the answering state. Differences in student activity between these three states were analyzed using a general linear model, 1s fNIRS data clips, and features, including averages from the three states, were classified using machine learning classification models such as support vector machines and k-nearest neighbor. The results show that the resting state is more active than learning in the dorsolateral prefrontal cortex, while answering questions is the most active of the three states in the entire PFC, and k-nearest neighbor achieves 98.5% classification accuracy for 1s fNIRS data. The results clarify the differences in PFC activity between resting, learning, and question-answering states in online learning scenarios and support the feasibility of developing an online learning assessment system using fNIRS and machine learning techniques.

Adaptive RPL Routing Optimization Model for Multimedia Data Transmission using IOT

Objectives: The main objectives of this research endeavor encompass the development of the Adaptive RPL Optimization (ARPLO) model to enhance data transmission efficiency within IoT networks. This includes constructing a grid-based network structure optimized for data transfer, selecting the most suitable nodes as grid head nodes to maximize network lifespan while minimizing energy consumption, implementing an innovative objective function-driven approach to optimize parent node selection, and integrating an Adaptive Deep Neural Network (ADNN) to accurately classify medical data. Methods: The research methodology entails several key steps. A grid-based network structure is established with IoT nodes and root nodes, where a DODAG process incorporating DIO messages is utilized for node ranking. To enhance energy efficiency, the Trickle algorithm is employed for control message optimization. Grid head nodes are chosen based on metrics such as root node fairness, residual energy, and load influence index. The novel Middle Order Optimal Routing (MOOR) objective function is utilized to optimize routing decisions. ADNN is implemented for precise medical data classification. The proposed model's performance is evaluated through simulation in a Python environment. Findings: The research findings demonstrate that the ARPLO model yields notable benefits compared to existing models. It achieves higher energy efficiency, improved throughput, enhanced packet delivery ratio (PDR), and an extended network lifespan. The Trickle algorithm contributes to efficient control message optimization. The MOOR-based routing approach improves multimedia medical data transfer. Moreover, the integration of ADNN enhances the accuracy of data classification, particularly in healthcare applications. The research outcomes align with the broader field's existing values and reports while offering novel insights that contribute to enhancing the existing knowledge base. ARPLO protocol performance reveals that there is increase of throughput of 31.2%, PDR by 7.12%, lifetime of 10.7 % with reduction of energy consumption by 12.72%, control overhead by 31.01% and end-to-end delay by 33.01%. Novelty: The novelty of this research lies in its comprehensive approach that integrates a grid-based network structure, MOOR-based optimization, and ADNN-based classification. The incorporation of the Trickle algorithm for energy-efficient communication is an innovative feature. The introduction of new metrics for grid head node selection, along with the application of the MOOR objective function for multimedia medical data routing, showcases the research's innovative contributions. Keywords: Internet of Things (IoT), RPL (Routing Protocol for Low­Power and Lossy Networks), Optimization, Routing, Multimedia, Healthcare

Tensile strength analysis of additively manufactured CM 247LC alloy specimen by employing machine learning classifiers.

Using a cutting-edge net-shape manufacturing technique called Additive Layer Manufacturing (ALM), highly complex components that are not achievable with conventional wrought and cast methods can be produced. As a result, the aerospace sector is paying closer attention to using this technology to fabricate superalloys based on nickel to develop the holistic gas turbine. Because of this, there is an increasing need for the mechanical characterisation of such material. Conventional mechanical testing is hampered by the limited availability of material that has been processed, especially given the large number of process factors that need to be assessed. Thus, the present study focuses on manufacturing CM247LC Ni-based superalloy with exceptional mechanical characteristics by laser powder bed fusion (L-PBF). This study evaluates the effect of input process variables such as laser power, scan speed, hatch distance and volumetric energy density on the mechanical performance of the LPBF CM247LC superalloy. The maximum value of as-built tensile strength obtained in the study is 997.81 MPa. Plotting Pearson's heatmap and the Feature importance (F-test) was used in the data analysis to examine the impact of input parameters on tensile strength. The accuracy of the tensile strength data classification by machine learning algorithms, such as k-nearest neighbours, Naïve Baiyes, Support vector machine, XGBoost, AdaBoost, Decision tree, Random forest, and logistic regression algorithms, was 92.5%, 83.75%, 83%, 85%, 87.5%, 90%, 91.25%, and 77.5%, respectively.

Dynamic Ensemble Selection for Imbalanced Data Streams With Concept Drift.

Ensemble learning, as a popular method to tackle concept drift in data stream, forms a combination of base classifiers according to their global performances. However, concept drift generally occurs in local data space, causing significantly different performances of a base classifier at different locations. Thus, employing global performance as a criterion to select base classifier is inappropriate. Moreover, data stream is often accompanied by class imbalance problem, which affects the classification accuracy of ensemble learning on minority instances. To drawback these problems, a dynamic ensemble selection for imbalanced data streams with concept drift (DES-ICD) is proposed. For data arrived in chunk-by-chunk, a novel synthetic minority oversampling technique with adaptive nearest neighbors (AnnSMOTE) is developed to generate new minority instances that conform to the new concept. Following that, DES-ICD creates a base classifier on newly arrived data chunk balanced by AnnSMOTE and merges it with historical base classifiers to form a candidate classifier pool. For each query instance, the optimal combination is constructed in terms of the performance of candidate classifiers in its neighborhood. Experimental results for nine synthetic and five real-world datasets show that the proposed method outperforms seven comparative methods on classification accuracy and tracks new concepts in an imbalanced data stream more preciously.

Heart Disease Diagnosis Using Decision Trees with Feature Selection Method

The advancement in treating medical data grows significantly daily. An accurate data classification model can help determine patient disease and diagnose disease severity in the medical domain, thus easing doctors' treatment burdens. Nonetheless, medical data analysis presents challenges due to uncertainty, the correlations between various measurements, and the high dimensionality of the data. These challenges burden statistical classification models. Machine Learning (ML) and data mining approaches have proven effective in recent years in gaining a deeper understanding of the importance of these aspects. This research adopts a well-known supervised learning classification model named a Decision Tree (DT). DT is a typical tree structure consisting of a central node, connected branches, and internal and terminal nodes. In each node, we have a decision to be made, such as in a rule-based system. This type of model helps researchers and physicians better diagnose a disease. To reduce the complexity of the proposed DT, we explored using the Feature Selection (FS) method to design a simpler diagnosis model with fewer factors. This concept will help reduce the data collection stage. A comparative analysis has been conducted between the developed DT and other various ML models, such as Logistic Regression (LR), Support Vector Machine (SVM), and Gaussian Naive Bayes (GNB), to demonstrate the effectiveness of the developed model. The results of the DT model establish a notable accuracy of 93.78\% and an ROC value of 0.94, which beats other compared algorithms. The developed DT model provided promising results and can help diagnose heart disease

Optimizing ant colony system algorithm with rule-based data classification for smart aquaculture

<span>Aquaculture is one of many industries where the use of artificial intelligence (AI) techniques has increased dramatically in recent years. Internet of things (IoT), AI, and big data are just a few of the technologies being used in smart aquaculture to increase productivity, efficiency, and system sustainability of aquaculture systems. Data classification, which involves finding patterns and relationships in huge datasets, is one of the most important tasks in smart aquaculture. The ant colony system (ACS) has been used to solve a number of optimization issues, including data classification. To provide a more practical and successful solution, this study proposes an improved ACS algorithm for rule-based data classification in smart aquaculture. The proposed algorithm combines the advantages of ACS and rule-based classification to optimize the number of rules and accuracy. The experimental results showed that the proposed algorithm outperformed the traditional AntMiner algorithm in terms of the number of rules and accuracy. The improved pheromone update technique could potentially increase data classification accuracy and convergence in smart aquaculture systems.</span>