Classification Of Attributes Research Articles

Cyber Attacker Profiling for Risk Analysis Based on Machine Learning.

The notion of the attacker profile is often used in risk analysis tasks such as cyber attack forecasting, security incident investigations and security decision support. The attacker profile is a set of attributes characterising an attacker and their behaviour. This paper analyzes the research in the area of attacker modelling and presents the analysis results as a classification of attacker models, attributes and risk analysis techniques that are used to construct the attacker models. The authors introduce a formal two-level attacker model that consists of high-level attributes calculated using low-level attributes that are in turn calculated on the basis of the raw security data. To specify the low-level attributes, the authors performed a series of experiments with datasets of attacks. Firstly, the requirements of the datasets for the experiments were specified in order to select the appropriate datasets, and, afterwards, the applicability of the attributes formed on the basis of such nominal parameters as bash commands and event logs to calculate high-level attributes was evaluated. The results allow us to conclude that attack team profiles can be differentiated using nominal parameters such as bash history logs. At the same time, accurate attacker profiling requires the extension of the low-level attributes list.

Seismic Avo Attributes and Machine Learning Techniques Characterise a Distributed Carbonate Build-Up Deposit System in the Salawati Basin, Eastern Indonesia

There are several undeveloped discoveries and drill-ready prospects in the Salawati Basin, Eastern Indonesia, especially offshore. These offer significant growth potential, particularly the challenging Miocene hydrocarbon-producing reservoirs. The seismic data suggests that this play extends to the north of the producing area, but this has not been confirmed by a successful well. The combination of standard seismic attributes with seismic Amplitude Variation with Offset (AVO) attributes is key to revealing the reservoir in the exploration phase. In this project, poststack attributes from an AV O inversion were used as input for an unsupervised clustering technique based on a Growing Neural Gas algorithm, to generate the most probable facies distribution as well as the probability per facies, in order to better characterise a complex regional channel deposition system. The classification of AVO-related seismic attributes as direct hydrocarbon indicators is used to extrapolate reservoir information from the seismic data that correlates with well data from surrounding fields. The study demonstrates that seismic volume-based unsuper-vised facies classification associated with advanced visualisation and detection helps delineate the prospect’s potential. In this example, the workflow identified a reservoir within the Kais interval of the Miocene Carbonate. The model also shows lateral variations in other reservoir intervals and contributes to the exploration hydrocarbon strategy.

Equitable modelling of brain imaging by counterfactual augmentation with morphologically constrained 3D deep generative models.

We describe CounterSynth, a conditional generative model of diffeomorphic deformations that induce label-driven, biologically plausible changes in volumetric brain images. The model is intended to synthesise counterfactual training data augmentations for downstream discriminative modelling tasks where fidelity is limited by data imbalance, distributional instability, confounding, or underspecification, and exhibits inequitable performance across distinct subpopulations. Focusing on demographic attributes, we evaluate the quality of synthesised counterfactuals with voxel-based morphometry, classification and regression of the conditioning attributes, and the Fréchet inception distance. Examining downstream discriminative performance in the context of engineered demographic imbalance and confounding, we use UK Biobank and OASIS magnetic resonance imaging data to benchmark CounterSynth augmentation against current solutions to these problems. We achieve state-of-the-art improvements, both in overall fidelity and equity. The source code for CounterSynth is available at https://github.com/guilherme-pombo/CounterSynth.

Text2Simulate: A Scientific Knowledge Visualization Technique for Generating Visual Simulations from Textual Knowledge

Recent research has developed knowledge visualization techniques for generating interactive visualizations from textual knowledge. However, when applied, these techniques do not generate precise semantic visual representations, which is imperative for domains that require an accurate visual representation of spatial attributes and relationships between objects of discourse in explicit knowledge. Therefore, this work presents a Text-to-Simulation Knowledge Visualization (TSKV) technique for generating visual simulations from domain knowledge by developing a rule-based classifier to improve natural language processing, and a Spatial Ordering (SO) algorithm to solve the identified challenge. A system architecture was developed to structurally model the components of the TSKV technique and implemented using a Knowledge Visualization application called ‘Text2Simulate’. A quantitative evaluation of the application was carried out to test for accuracy using modified existing information visualization evaluation criteria. Object Inclusion (OI), Object-Attribute Visibility (OAV), Relative Positioning (RP), and Exact Visual Representation (EVR) criteria were modified to include Object’s Motion (OM) metric for quantitative evaluation of generated visual simulations. Evaluation for accuracy on generated simulation results were 90.1, 84.0, 90.1, 90.0, and 96.0% for OI, OAV, OM, RP, and EVR criteria respectively. User evaluation was conducted to measure system effectiveness and user satisfaction which showed that all the participants were satisfied well above average. These generated results showed an improved semantic quality of visualized knowledge due to the improved classification of spatial attributes and relationships from textual knowledge. This technique could be adopted during the development of electronic learning applications for improved understanding and desirable actions.

Automatic simplification of the geometry of a cartographic line using contractive self-mapping – illustrated with an example of a polyline band

Abstract The present article is another attempt to adapt map geometry to automatic digital cartography. The paper presents a method of digital polyline generalisation that uses contractive self-mapping. It is a method of simplification, not just an algorithm for simplification. This method in its 1996 version obtained a patent entitled “Method of Eliminating Points in the Process of Numerical Cartographic Generalisation” – Patent Office of the Republic of Poland, No. 181014, 1996. The first results of research conducted using the presented method, with clearly defined data (without singular points of their geometry), were published in the works of the authors in 2021 and 2022. This article presents a transition from the DLM (Digital Landscape Model) to the DCM (Digital Cartographic Model). It demonstrates an algorithm with independent solutions for the band axis and both its edges. The presented example was performed for the so-called polyline band, which can represent real topographic linear objects such as rivers and boundaries of closed areas (buildings, lakes, etc.). An unambiguous representation of both edges of the band is its axis, represented in DLM, which can be simplified to any scale. A direct consequence of this simplification is the shape of the band representing the actual shape of both edges of the object that is classified in the database as a linear object in DCM. The article presents an example performed for the so-called polyline band, which represents real topographic linear objects (roads, rivers) and area boundaries. The proposed method fulfils the following conditions: the Lipschitz condition, the Cauchy condition, the Banach theorem, and the Salichtchev’s standard for object recognition on the map. The presented method is objective in contrast to the previously used approximate methods, such as generalisations that use graph theory and fractal geometry, line smoothing and simplification algorithms, statistical methods with classification of object attributes, artificial intelligence, etc. The presented method for changing the geometry of objects by any scale of the map is 100% automatic, repeatable, and objective; that is, it does not require a cartographer’s intervention.

Historical investigation on paradigm of "species standardization" of Chinese herbal medicine

The paradigm can be summarized as the same methodological principles followed by the members of the learning community who received common education, training, and edification in the common social and historical stage, and had common academic views, concepts, and terms. The "species standardization" paradigm of botanical drugs in China has experienced the period of non-standardization of names(the end of the Warring States period), the period of multi-dimensional textual research paradigm of traditional materia medica(from the end of the Warring States period to 1930s), and periods of modern plant taxonomy system research paradigm(from 1930s to the present). The multi-dimensional textual research paradigm of traditional materia medica is characterized by the textual research of species in different regions and years based on the classification of the natural attributes and the habit of correspondence between names and materials, from the multi-dimensional investigation of the name, appearance, color, smell, texture, growth environment, efficacy, and morphological characteristics of medicinal materials, to find the correspondence between names and materials. The purpose of the progress in traditional species standardization is practicality for humans, and there is a lack of systematic research on the classification and naming principles in line with its laws from the plant itself, resulting in inaccurate research results. The research paradigm of modern plant taxonomy is herbal research. The characteristics of this paradigm are using plant taxonomy knowledge to conduct textual research to confirm the species of the drug and calibrate the scientific name by the double name method, based on the plant morphology, harvest season, origin distribution, ecological habits, names, drug efficacy, and other data in the lite-rature of previous dynasties. This paradigm combines the knowledge of traditional materia medica and modern plant taxonomy, and the results of species research are accurate.

Classification And Regression Tree (CART) based resource allocation scheme for Wireless Sensor Networks

The 21st century is the era of smart sensors, intelligent computations, and communication technologies. Wireless Sensor Networks (WSN) play a vital role in performing many remote applications without human intervention. WSNs are an integral part of an ubiquitous computing and Internet of Everything (IoE). A WSN is a decentralized network comprised of several tiny but powerful sensor nodes. Sensor nodes are restricted in terms of battery life, communication range, bandwidth, processing latency, and memory. Effective usage of WSN resources is a challenging task to enhance network lifespan, increase throughput, reduce computational delay and minimize control overheads. Several intelligent strategies are proposed to improve WSN performance and enhance the lifespan of the network by adopting intelligent resource management schemes. In WSN, effective and intelligent resource management involves resource discovery, resource scheduling, and resource allocation. In this paper, a Classification and Regression Tree (CART) supervised machine learning algorithm is used to deal with incomplete information about the network i.e., uncertainty and dynamicity for effective resource allocation. The scheme operates as follows: The k-means clustering algorithm is applied to the network, clusters are formed, the Cluster Head (CH) is selected, and the k-NN algorithm is applied to find the number of neighbor nodes i.e., Cluster Members (CM) in the cluster. The attributes of CH like distance from base station, degree of connectivity, congestion rate, data type and size aggregated at CH after performing a task and channel quality are calculated. Aggregation and classification of CM and CH attributes (data sets) use intelligent search and feature selection algorithms. The data sets are then processed for the training and prediction phases. A decision tree model is built using target attribute, which is resource (bandwidth) allocation. A heat map and confusion matrix are generated and a performance evaluation of the proposed scheme is done. Simulation results show that the performance of the proposed CART based resource allocation approach is better as compared with the “Linear Regression (LR), Iterative Dichotomiser 3 (ID3), and Neural Network (NN)” schemes in terms of resource allocation accuracy, allocation computational delay, data transmission efficiency, etc.

태권도공연 제작 과정의 효율성 연구

The purpose of this study is to increase the efficiency of the production process of creators of Taekwondo performance works. In order to achieve the purpose, the data were analyzed with an in-depth interview technique to 9 experts and the following conclusions were obtained. First, as a result of classification and analysis of selective attributes in the production process of Taekwondo performance, it consists of seven upper categories: work completion, directing direction, taekwondo characteristics, directing ability, work characteristics, characterization, time, and 18 subcategories; Narrative, atmosphere, speed, time and space, futility, harmony, uniformity, communication, guidance, understanding, creativity, nuance of technical movement, expression method, style, practice time, and work time were derived. Second, it is believed that the results of the categorized selection attribute content analysis can be used as basic data for the production process.

Classification of Attributes on Green Manufacturing Practices: A Systematic Review

Classification of Attributes on Green Manufacturing Practices: A Systematic Review

Construction of a Generic and Evolutive Wheel and Lexicon of Food Textures.

In the context of data management and processing, food science needs tools to organize the results of diverse studies to make the data reusable. In sensory analysis, there are no classification or wheel of textural attributes that can be used to interpret the results of sensory studies. Research from the literature and databases was used to elaborate a list of attributes related to texture. With the help of a group of experts in food texture, work on these attributes and the related concepts was conducted to classify them into several categories, including intensity levels. The classification was represented as a texture wheel, completed by a generic lexicon of definitions of texture concepts. The work can be useful as a reference in texture attributes related to foods, and thanks to implementation in a general ontology based on food processing and observation, it can help query and interpret texture-related results from sensory studies.

A computationally efficient learning model to classify audio signal attributes

<p>The era of machine learning has opened up groundbreaking realities and opportunities in the field of medical diagnosis. However, it is also observed that faster and proper diagnosis of any diseases/medical conditions require proper analysis and classification of digital signal data. It indicates the proper identification of tumors in the brain. Brain magnetic resonance imaging (MRI) data has to be appropriately classified, and similarly, pulse signal analysis is required to evaluate the human heart operating condition. Several studies have used machine learning (ML) modeling to classify speech signals, but very few studies have explored the classification of audio signal attributes in the context of intelligent healthcare monitoring. The study thereby aims to introduce novel mathematical modeling to analyze and classify synthetic pulse audio signal attributes with cost-effective computation. The numerical modeling is composed of several functional blocks where deep neural network-based learning (DNNL) plays a crucial role during the training phase, and also it is further combined with a recurrent structure of long-short term memory (R-LSTM) feedback connections (FCs). The design approaches further experiment in a numerical computing environment in terms of accuracy and computational aspects. The classification outcome of the proposed approach shows that it attains approximately 85% accuracy, which is comparable to the baseline approaches and execution time.</p>

RETRACTED: Evaluation of biometric communication and authenticate recognition using ANN with PSO algorithm

This research investigates the novel techniques which provide the detailed information on the biometric images used along with the methods applied for biometric image pre-processing. It also describes the proposed methodology which was implemented with the method of optimized Particle Swarm Optimization (PSO) with Artificial Neural Network (ANN) algorithm for classification of attributes. In the current work, a big effort has been implemented for designing an efficient technique for recognizing the biometric images, especially for the modalities like finger print and retina image. Initially, the pre-processing module used the method of histogram equalization to enhance the contrasts of entire image in order to get the best image quality. This makes the image adaptable for further processing. Next, the feature extraction module has the involvement of two image sets (finger print and retina image). The Gray Level Co-occurrence Matrix (GLCM) was used for extracting the needed features in this module. Next is Feature Based Fusion Technique (FBFT) for reducing the features for authentication purpose. This research work uses the FBFT to get fused feature vector. Finally, deals with the non-recognition and recognition of the images. The images were tested by using Artificial Neural Network (ANN). Here, the recognition is done by ANN and the optimization is done by the sophisticated function of Particle Swarm Optimization Algorithm (PSOA). ANN does the classification of images as recognized and non-recognized and yields best results.

Attribute Feature Classification of English Grammar Entry Base Based on Support Vector Machine Classification Algorithm.

An attribute feature classification method of English grammar vocabulary entry database based on support vector machine classification algorithm is proposed; this method takes news English as the research object and focuses on the classification of attributes and features of the English grammar lexicon database. First, the k-means algorithm is used to cluster the training set, and the one-to-many method is used to train two types of classifiers for the texts that cannot be correctly clustered in each class, that is, the classifiers of the corresponding categories are trained, and then the training set passed through a pair of the classifier generated by multiple SVMs is tested, and the samples that fall in the inseparable area are retrained by a one-to-one method, so as to achieve the purpose of balancing the training samples and reducing the inseparable area. The results show that, compared with the FDAGSVM algorithm, the proposed three multiclass classification algorithms have significantly improved classification speed and classification accuracy, and the macro average accuracy rates are 77.94%, 73.94%, and 72.36%, respectively. While ensuring the classification speed and classification accuracy of the single-label samples, the multiclass classification is realized, and it has high accuracy, recall rate, and value, which better solves the multiclass classification problem and expands the classification capability of the support vector machine. In addition, a comprehensive index based on the SVM classification algorithm is proposed to ensure the specialization of the attribute feature classification.

ONLINE SHOPPING USER PORTRAIT MODEL CONSIDERING PSYCHOLOGICAL AND EMOTIONAL FACTORS

ONLINE SHOPPING USER PORTRAIT MODEL CONSIDERING PSYCHOLOGICAL AND EMOTIONAL FACTORS

Study of Classification of Quality Attributes of FoodCarvingEducation Utility through RevisedIPA

본 연구는 푸드카빙 교육을 받은 수강생들이 지각하는 푸드카빙 효용성은 어떠한지 살펴보고, 어떠한 효용성들이 있고, 효용성의 품질 속성은 어떠한지 살펴보고자 하였다. 이러한 연구목적을 규명하기 위하여 푸드카빙 교육을 받은 경험이 있는 수강생을 모집단으로 설정하였고, 설문조사는 2020년 7월 20일부터 8월 20일까지 한 달간 진행되었다. 설문지는 총 500부를 배부하였고, 435부가 회수되었고, 불성실한 설문지 38부를 제외하고 397부가 최종 분석에 사용되었다. 분석결과, 중요 실행 요소로는 푸드카빙 교육은 지루하지 않고, 남녀노소 할 수 있고, 잡생각이 사라지고, 자존감이 높아지고, 자신의 가치가 높아지고, 마음이 편해지고, 우울함이 사라지는 등의 수업 자체에서 오는 만족감의 속성들이 포함되었다. 매력적 요소에는 푸드카빙 기술이 수익으로 연결되고, 강좌를 개설하고, 적은 금액으로 공방을 오픈하는 등의 수익 측면과 관련된 항목들이 포함되었다. 아울러, 기본적 요소에는 집중할 수 있고, 기분이 좋아지고, 자격증을 취득하고, 전문가가 될 수 있다는 측면의 항목들이 포함되었다. 이는 대체로 푸드카빙을 처음 배우는 수강생들이 기본적으로 가지는 목적들이 포함되었다.

Structural Equation Modeling (SEM) and Temporal Dominance of Sensations (TDS) in the Evaluation of DOC Douro Red Wine’s Sensory Profile

In the Portuguese Douro region, several DOC (Denomination of Controlled Origin) Douro red wines are produced and, due to the peculiar characteristics of the three Douro sub-regions, present particular imprinted terroirs, that can be perceived when tasted. Considering the DOC Douro wine’s sensory profile and terroir, this study aimed to analyze the sensory characteristics of red wines produced in the three Douro sub-regions (Baixo Corgo, Cima Corgo, and Douro Superior) by a single point sensory technique, a Quantitative Descriptive Analysis—QDA® and also applying a temporal method-TDS (Temporal Dominance of Sensations). The use of QDA and TDS methods proved to be efficient in the wine’s sensory profile characterizing. The QDA® method allowed a detailed classification of attributes; however, the TDS method proved to be much more efficient. Moreover, the wines of the three sub-regions presented profiles with characteristics very similar in olfactory and taste/flavor aspects, pointing out a huge relation between the characteristics of the three sub-regions and the grape varieties present in the wines. Globally, the olfactory profile of wines is characterized by Fruity, Floral, and Balsamic aromatic notes, while the taste/flavor profile stands out, highlighting Astringency and Acidity and, again, Fruity as the main in-mouth aroma. It was also possible to conclude that TDS is a fast method that is easy to apply and has excellent results in the evaluation of the olfactory and taste/flavor profile of wines and, with a larger set of samples, it would be possible to obtain characteristic TDS curves for each Douro sub-region, providing a wine’s fingerprint that could be used for authentication and traceability purposes.

Deep Multi-Task Multi-Label CNN for Effective Facial Attribute Classification

Facial Attribute Classification (FAC) has attracted increasing attention in computer vision and pattern recognition. However, state-of-the-art FAC methods perform face detection/alignment and FAC independently. The inherent dependencies between these tasks are not fully exploited. In addition, most methods predict all facial attributes using the same CNN network architecture, which ignores the different learning complexities of facial attributes. To address the above problems, we propose a novel deep multi-task multi-label CNN, termed DMM-CNN, for effective FAC. Specifically, DMM-CNN jointly optimizes two closely-related tasks (i.e., facial landmark detection and FAC) to improve the performance of FAC by taking advantage of multi-task learning. To deal with the diverse learning complexities of facial attributes, we divide the attributes into two groups: objective attributes and subjective attributes. Two different network architectures are respectively designed to extract features for two groups of attributes, and a novel dynamic weighting scheme is proposed to automatically assign the loss weight to each facial attribute during training. Furthermore, an adaptive thresholding strategy is developed to effectively alleviate the problem of class imbalance for multi-label learning. Experimental results on the challenging CelebA and LFWA datasets show the superiority of the proposed DMM-CNN method compared with several state-of-the-art FAC methods.

Machine learning models for classification and identification of significant attributes to detect type 2 diabetes

Type 2 Diabetes (T2D) is a chronic disease characterized by abnormally high blood glucose levels due to insulin resistance and reduced pancreatic insulin production. The challenge of this work is to identify T2D-associated features that can distinguish T2D sub-types for prognosis and treatment purposes. We thus employed machine learning (ML) techniques to categorize T2D patients using data from the Pima Indian Diabetes Dataset from the Kaggle ML repository. After data preprocessing, several feature selection techniques were used to extract feature subsets, and a range of classification techniques were used to analyze these. We then compared the derived classification results to identify the best classifiers by considering accuracy, kappa statistics, area under the receiver operating characteristic (AUROC), sensitivity, specificity, and logarithmic loss (logloss). To evaluate the performance of different classifiers, we investigated their outcomes using the summary statistics with a resampling distribution. Therefore, Generalized Boosted Regression modeling showed the highest accuracy (90.91%), followed by kappa statistics (78.77%) and specificity (85.19%). In addition, Sparse Distance Weighted Discrimination, Generalized Additive Model using LOESS and Boosted Generalized Additive Models also gave the maximum sensitivity (100%), highest AUROC (95.26%) and lowest logarithmic loss (30.98%) respectively. Notably, the Generalized Additive Model using LOESS was the top-ranked algorithm according to non-parametric Friedman testing. Of the features identified by these machine learning models, glucose levels, body mass index, diabetes pedigree function, and age were consistently identified as the best and most frequently accurate outcome predictors. These results indicate the utility of ML methods in constructing improved prediction models for T2D and successfully identified outcome predictors for this Pima Indian population.

Проблема выявления источника (атрибуции) кибератак – фактор международной безопасности

The work is devoted to the problem of attribution of cyberattacks as one of the top important issues on the global agenda of world politics. The subject of this research is targeted offensive computer attacks that perform state tasks. The authors have systematically analyzed the conceptual apparatus, classification features and general indicators, a set of problems related to identification of reliable sources of cyberattacks, as well as the features of targeted attacks and cybernetic groups supported by state actors. This allowed to conclude that the problem of attribution is emerging as a new scientific direction in the field of international information security. Based on the analysis of relevant studies of academic institutions and private companies, an attribution classification was proposed. It was utilized to review conceptual informal models of targeted computer attacks. The authors have analyzed issues of the applied methodological apparatus, which may affect the reliability of conclusions on imposing responsibility on the perpetrator of cyber-attack, while noting that there is a contradiction between the perception of the attribution results by various political blocs and unions. At the same moment, the USA and its allies are reported to pursue a policy of dominance in the field of attribution of targeted cyber-attacks. The paper proves the impact of modern breakthrough information technologies of the Fourth Industrial Revolution on the security of cyberspace as well as the impact of the attribution problem on the level of international security and strategic stability. This issue includes a complex set of political, regulatory, organizational and technical tasks with a high degree of uncertainty, where political aspects are playing a central role. The conclusions are confirmed by the statistics of public reports of IT companies and publications of leading research institutes. The main scientific result of the work is formulation of the problem of divergence in the attribution of computer attacks between political blocs and alliances, which negatively affects international security. The authors propose topical measures to solve the problem of attribution of cyberattacks.

Identification of cocoa bean quality by near infrared spectroscopy and multivariate modeling

Cocoa is a commodity responsible for the income of millions of people and the manufacture of several important products for the food, pharmaceutical, and cosmetic industries. Its quality is associated with several factors involved in the processing steps, mainly in fermentation and drying. The objective of this study was to evaluate the application of near-infrared spectroscopic data associated with multivariate analysis to classify cocoa beans according to their quality and predict attributes such as pH and total acidity by PLS-DA and PLS, respectively. The pH values (4.4-6.7) and total acidity (6.12-29.9) were determined by conventional methods. The PLS-DA proved to be effective in differentiating the classes of cocoa samples with superior and inferior quality, presenting in the validation 100% and 71.43% correct cocoa bean classification with inferior Quality and Higher Quality, respectively. The models obtained by PLS presented satisfactory parameters, being classified as having moderate practical utility and excellent predictive capacity for pH and moderate practical utility and reasonable predictive capacity for total acidity. Thus, the potential of the NIRS technology associated with chemometrics was found and showed efficiency in the classification and prediction of attributes in cocoa beans.