Linear Classification Methods Research Articles

Fusion regression methods with repeated functional data

Linear regression and classification methods with repeated functional data are considered. For each statistical unit in the sample, a real-valued parameter is observed over time under different conditions related by some neighborhood structure (spatial, group, etc.). Two regression methods based on fusion penalties are proposed to consider the dependence induced by this structure. These methods aim to obtain parsimonious coefficient regression functions, by determining if close conditions are associated with common regression coefficient functions. The first method is a generalization to functional data of the variable fusion methodology based on the 1-nearest neighbor. The second one relies on the group fusion lasso penalty which assumes some grouping structure of conditions and allows for homogeneity among the regression coefficient functions within groups. Numerical simulations and an application of electroencephalography data are presented.

THE CORRELATIVE VARIABILITY OF THE FINAL SCORE BY LINEAR CLASSIFICATION AND LIFETIME PRODUCTIVITY INDICATORS AMONG BLACK-AND-WHITE DAIRY COWS OF DIFFERENT BREEDS

A study on the correlative variability of the final score by linear classification and lifetime productivity indicators of Black-and-White Dairy cows of different breeds was conducted in the herd of the private enterprise "Burynske" in the Podlisniv branch of the Stepaniv territorial community. Foreign studies on the cows evaluation of different breeds by conformation type using the linear classification method were analyzed, which proved that the lifetime productivity indicators of animals depend on many genotypic and paratypic factors. In this direction, the existence of significant moderate and close positive correlations between the majority of linear traits and functional life of cows in the herd was established. It has also been proven that the final score level of cows by type is closely correlated with descriptive traits that significantly affect lifetime productivity indicators. According to the results of our own research, it was found that degree of the final score of first-born cows for the conformation type directly determines the variability level of indicators to the duration of use and lifetime productivity within both experimental breeds. Cows of the Holstein and Ukrainian Black-and-White dairy breeds with a rating of "very good" exceeded animals with ratings of "good plus" and "good" by indicators of duration productive use and milk productivity throughout life. Holstein cows rated "very good" were better than animals rated "good plus" and "good" in terms of lifetime milk yield and milk fat yield, respectively by 4589 and 21653 kg, and 191 and 918 kg. The final score, in the presence of a correlation with lifetime productivity indicators, can serve as a predictor of longevity, which will allow effective selection of first-born dairy cows based on the results of a linear evaluation in the first months of lactation.

Physical fitness and motor competence performance characteristics of Chinese elite youth athletes from four track and field throwing disciplines-a cross-sectional study.

Purpose: For systematic athletic training and targeted talent development, it is essential to know the physical fitness and motor competencies of top athletes in detail. However, it can be difficult to identify differences in performance requirements and thus to provide adequate support, especially for sports that at first glance appear to have similar demands-such as track and field throwing disciplines. Therefore, the aim of the study was to examine the physical fitness and motor competence of top athletes from different throwing disciplines and to check whether the athletes' performance parameters match the specific requirements of the respective sport. Methods: The study involved 289 male youth athletes (aged 14-18years) across four distinct throwing disciplines: shot put (n = 101), hammer throw (n = 16), discus throw (n = 63), and javelin throw (n = 109). The performance evaluation comprised three anthropometric measurements and twelve motor performance prerequisites applicable to the throwing disciplines. Discriminant analysis and neural networks (Multilayer Perceptron) were implemented to determine the possibility of distinguishing among athletes from the four sports. Results: The study's findings indicate that in male throwing athletes, disparities in general physical fitness and motor proficiency assessments discern the majority of talented young athletes based on their specific sport (discriminant analysis: 68.2%; multilayer perceptron analysis: 72.2%). This remains applicable irrespective of the classification method employed. Discus throwers possessed a height advantage, while shot putters and hammer throwers exhibited superior arm strength. Javelin throwers displayed better explosive strength and sprinting speed. Except for the hammer throwers, all events demonstrated a high level of explosive power in the medicine ball forward or backward throw test, which was especially crucial for shot put and discus athletes. Conclusion: The significance of physical fitness and motor competence tests in identifying and transferring talented athletes in track and field throwing disciplines has been affirmed. Using linear and non-linear classification methods, most athletes could be assigned to their correct sport. However, this also shows that slightly different training and talent identification is required for each of these sports. Furthermore, non-linear analysis methods can provide useful support for the development processes in junior competitive sports.

Research of methods based on neural networks for the analysis of the tonality of the corps of the texts

The object of the study is methods based on neural networks for analyzing the tonality of a corpus of texts. To achieve the goal set in the work, it is necessary to solve the following tasks: study the theoretical material for learning deep neural networks and their features in relation to natural language processing; study the documentation of the Tensorflow library; develop models of convolutional and recurrent neural networks; to develop the implementation of linear and non-linear classification methods on bag of words and Word2Vec models; to compare the accuracy and other quality indicators of implemented neural network models with classical methods. Tensorboard is used for learning visualization. The work shows the superiority of classifiers based on deep neural networks over classical classification methods, even if the Word2Vec model is used for vector representations of words. The model of recurrent neural network with LSTM blocks has the highest accuracy for this corpus of texts.

Style linear k-nearest neighbor classification method

Most studies on linear k-nearest neighbor methods assume independent and identically distributed training and testing samples. However, this is not the case in many practical applications, particularly when each class shares a homogeneous style. We propose a novel style linear k-nearest neighbor method to achieve two goals: (a) the linear k-nearest neighbor method should mine the stylistic data for explicit or implicit stylistic features while obtaining linear expressions and effectively transfer them to the predictor through matrix expressions; (b) the similarity of stylistic features between testing and training samples should be numerically quantified to enhance the generalizability of the predictor. To this end, firstly we introduce style matrices to express the style information of each class. In order to prevent the style matrices from degenerating into identity matrices, we introduce enhanced nodes to separate the manifold structure of the original data. The dual representation of matrix and enhanced nodes makes it easy to extract stylistic features from data. Furthermore, we introduce the style membership vector for the first time in the linear k-nearest neighbor method to calculate the style similarity of testing samples to each class, and determine the labels of testing samples easily and accurately through the style membership vector. We also propose an implementable alternating optimization strategy for the proposed method, which decomposes the complex optimization problem into independent subproblems for easy implementation. The experimental results demonstrate that the proposed method keeps comparable generalization capability on the total 9 ordinary datasets. And it outperforms the comparative methods, and even achieving 9.76 % improvement of average testing accuracy over the state-of-the-art weighted locally linear k-nearest neighbor method on the total 6 stylistic datasets.

Features of conformation type of black-andwhite cows of different origin estimated by linear classification method

The research was carried out in the aspect of studying features of conformation type cows of Ukrainian Black-and-White dairy and Holstein breeds estimated by linear classification method. Basis for research was the herd of company "Ukrlandfarming" private enterprise "Burynske", Pidlisniv branch in Sumy region. The aim was to conduct a comparative analysis of experimental breeds according to the assessment of linear classification and determine the impact of final score for type on milk productivity first-born cows. In general, first-born cows of both breeds were characterized by good linear estimation, especially on a 100-score scale. However, animals of Holstein breed in comparison with Ukrainian Black-and-White dairy were better in terms of body parts, which characterize the dairy type by 1.3 score, body - by 0.8 score, udder - by 1.2 score and the final assessment - by 0.8 score (P <0.001). On the trait of chest width, Holstein animals were inferior to their Ukrainian Black-and-White dairy breeds with reliable difference of 1.3 score, but exceeded in angularity - by 1.2 score, strength of fore udder parts attachment - on 0.8 score, central ligament - by 1.1 score (P <0.001). The final estimation level of first-born cows for conformation type directly proportionally determined their milk productivity in 305 days of the first lactation. Cows of Holstein and Ukrainian Black-andWhite dairy breed with score - "very good" exceeded their peers with score - "good with plus" for milking by 785 and 589 kg (P˂0.01) respectively, and with score of "good" - by 1705 and 1502 kg (P˂0.001). The use in the selection process of dairy cattle, in determining cows breeding value, linear classification method will be a very effective means of objective definition breed features of conformation type. The presence of correlative relationship between the final score and milk productivity level will contribute to the effectiveness of selection in the indirect sampling of animals on these traits. Key words: Holstein, Ukrainian Black-andWhite dairy, linear type assessment, conformation, milk yield.

Linear type traits that characterize body development as the predictors of lifetime of Ukrainian Black-and-White dairy and Holstein cows

The researches were conducted in the aspect of studying the problem of dairy cows lifetime depending on the linear traits assessment of the conformation type. By the linear classification method were estimated firstborn cows of the Ukrainian Black-and-White dairy (UBWD) and Holstein (H) breeds. Descriptive traits of the conformation that characterize the overall body structure were studied: chest width, body depth, angularity, body condition, rear width and position. According to the research results, the relative variability of the descriptive type traits assessment with cow's lifetime was determined. The relative variability of chest width and cow's lifetime was curvilinear. Cows for chest width assessment of 4–7 score had a higher lifespan with a variability of 2704–2844 (UBWD) and 2688–2789 (H) days. The highest lifetime was in animals with an assessment for body depth development of 6-9 score with an unreliable predominance of Ukrainian Black-and-White dairy breed cows, with the highest grades of both breeds 2824 (UBWD) and 2802 (H) days with an assessment of 9 score. Animals with an optimal angularity estimation of 5 score differed, lived the longest – 2842 (UBWD) and 2828 (H) days, while with increasing and decreasing score for this trait, the number of lifetime days of cows declined. According to the rump width assessment, lifespan was highest in cows with a 9 score – 2766 days (UBWD) and 2832 days (H). Cows of UBWD with the highest assessment for this body part development of 9 score were used on 592 (P <0.001), and Holstein on 708 (P <0.001) days longer compared to animals with 1 score. The maximum lifetime of animals with an average assessment for body condition of 5 score was 2842 (UBWD) and 2774 (H) days. Cows with body condition score below average lived and were used in the herd much longer than with a higher one. The degree of relationship variability between the assessment of these traits and the lifetime of animals depended on the specific body part. Key words: Ukrainian Black-and-White dairy breed, Holstein, linear type traits, lifetime.

Development of a Flexible Smart Wearable Oximeter Insole for Monitoring SpO2 Levels of Diabetics’ Foot Ulcer

This work presents the development of a fully functional prototype of a wearable smart shoe insole that can monitor arterial oxygen saturation (SpO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> ) levels at the foot of a diabetic patient using photoplethysmography (PPG) signals. Continuous monitoring of SpO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> levels at foot in patients with diabetic foot ulcer (DFU) can provide critical information on the severity of the ulcer, the wound healing process, and alerting clinicians for critical limb ischemia. The developed oximetry system seamlessly integrates the Internet of things (IoT) via a custom-developed Android mobile application, thus enabling “at-home” monitoring. Twenty healthy subjects were tested, and the insole oximeter was able to successfully estimate SpO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> levels at the toe. An average error of ≈ 2.6% was calculated for the measured/estimated SpO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> levels at the subjects’ toe when compared to a reference oximeter attached to the finger. Perfusion Index (PI) - which represents the blood flow/supply in tissues - was used as a method for validating the oximeter readings. It was observed that fingers (index fingers) generally have larger PI values when compared to the toe, while PI results of both monitoring sites were in the acceptable range. In addition, a dataset was formed for the twenty test subjects, and machine learning (ML) techniques were applied to predict the SpO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> level and the site of measurement (Finger or Toe), using multiple linear regression and classification methods. The ML results show that AC components of the PPG signals have a more significant contribution to the SpO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> estimations when compared to the DC components. In addition, KNN (K=1) classifier, was able to successfully predict the monitoring sites, with a test accuracy of 96.86%.

Recognition of Continuous Face Occlusion Based on Block Permutation by Using Linear Regression Classification

Face occlusion is still a key issue in the study of face recognition. Continuous occlusion affects the overall features and contour structure of a face, which brings significant challenges to face recognition. In previous studies, although the Representation-Based Classification Method (RBCM) can better capture the differences in different categories of faces and accurately identify human face images with changes in light and facial expressions, it is easily affected by continuous occlusion. For face recognition, there is a situation where face error recognition occurs. The RBCM method frequently learns to cover the characteristics of face recognition and then handle face error recognition. Therefore, the elimination of occlusion information from the image is necessary to improve the robustness of such models. The Block Permutation Linear Regression Classification (BPLRC) method proposed in this paper includes image block permutation and Linear Regression Classification (LRC). The LRC algorithm belongs to the category of nearest subspace classification and uses the Euclidean distance as a metric to classify images. The LRC algorithm is based on one of the classification methods that is susceptible to outliers. Therefore, block permutation was used with the aim of establishing an image set that does not contain much occlusion information and constructing a robust linear regression model. The BPLRC method first modulates all the images and then lists the schemes that arrange all segments, enters the image features of various schemes into linear models, and classifies the result according to the minimum residual of the person’s face image and reconstruction image. Compared to several state-of-the-art algorithms, the proposed method effectively solves the continuous occlusion problem for the Extended Yale B, ORL, and AR datasets. The proposed method recognizes the AR data concentration scarf to cover the accuracy of human face images to 93.67%. The dataset recognition speed is 0.094 s/piece. The arrangement method can be combined not only with the LRC algorithm, but also other algorithms with weak robustness. Due to the increase in the number of blocks and the increase in the calculation index of block arrangement methods, it is necessary to explore reasonable iteration methods in the future, quickly find the optimal or sub-best arrangement scheme, and reduce the calculation of the proposed method.

Implementation of Block-Level Double Encryption Based on Machine Learning Techniques for Attack Detection and Prevention

Cloud computing is one of the most important business models of modern information technology. It provides a minimum of various services to the user interaction and low cost (hardware and software). Cloud services are based on the newline architectures on virtualization by using the multitenancy for better resource management and newline strong isolation between several virtual machines (VMs). The spying on a victim VM is challenging, particularly when one wants to use per-core microarchitectural features as a side channel. For example, the cache contains the most potential for damaging side channels, but shared information across different cores affects the cloud information. To overcome this problem, propose the Secure Block-Level Double Encryption (SBLDE) algorithm for user signature verification in the cloud server. It uses identity-based detection techniques to monitor the colocated VMs to identify abnormal cache data and channel behaviors typically during VM data transformation. The identity-based linear classification (IBLC) method is used for classifying the attacker channel when the data is transferred/retrieved from the VM cloud server. This cloud controller finds the channel misbehavior to block the port or channel, changing other available ports’ communication. The service verification provides strong user access permission on the cloud server when the unknown request to the cloud server suddenly executes the key authentication to verify the user permission. This linear classification trains the existing side-channel attack datasets to the classifier and identifies the VM cloud’s attack channel. The study focused on preventing attacks from interrupting the system and serves as an effective means for cross-VM side-channel attacks. This proposed method protects the cloud data and prevents cross-VM channel attack detection efficiently, compared to other existing methods. In this overall proposed method, SBLDE’s performance is to be evaluated and then compared with the existing method.

Features of conformation type of black-andwhite cows of different origin estimated by linear classification method

The research was carried out in the aspect of studying features of conformation type cows of Ukrainian Black-and-White dairy and Holstein breeds estimated by linear classification method. As for the basis we took the herd of the company "Ukrlandfarming, private enterprise "Burynske", Pidlisniv branch in Sumy region. The aim was to conduct a comparative analysis of experimental breeds according to the assessment of linear classification and determine the impact of final score for type on milk productivity first-born cows. In general, first-born cows of both breeds were characterized by good linear estimation, especially on a 100-score scale. However, animals of Holstein breed in comparison with Ukrainian Black-and-White dairy were better in terms of body parts, which characterize the dairy type by 1.3 score, body - by 0.8 score, udder - by 1.2 score and the final assessment - by 0.8 score (P &lt;0.001).On the trait of chest width, Holstein animals were inferior to their Ukrainian Black-and-White dairy breeds with reliable difference of 1.3 score, but exceeded in angularity - by 1.2 score, strength of fore udder parts attachment - on 0.8 score, central ligament - by 1.1 score (P &lt;0.001).The final estimation level of first-born cows for conformation type directly proportionally determined their milk productivity in 305 days of the first lactation. Cows of Holstein and Ukrainian Black-and-White dairy breed with score - "very good" exceeded their peers with score - "good with plus" for milking by 785 and 589 kg (P˂0.01) respectively, and with score of "good" - by 1705 and 1502 kg (P˂0.001). The use in the selection process of dairy cattle, in determining cows breeding value, linear classification method will be a very effective means of objective definition breed features of conformation type. The presence of correlative relationship between the final score and milk productivity level will contribute to the effectiveness of selection in the indirect sampling of animals on these traits. Key words: Holstein, Ukrainian Black-and-White dairy, linear type assessment, conformation, milk yield.

Development and validation of predictive models for COVID-19 outcomes in a safety-net hospital population.

ObjectiveTo develop predictive models of coronavirus disease 2019 (COVID-19) outcomes, elucidate the influence of socioeconomic factors, and assess algorithmic racial fairness using a racially diverse patient population with high social needs.Materials and MethodsData included 7,102 patients with positive (RT-PCR) severe acute respiratory syndrome coronavirus 2 test at a safety-net system in Massachusetts. Linear and nonlinear classification methods were applied. A score based on a recurrent neural network and a transformer architecture was developed to capture the dynamic evolution of vital signs. Combined with patient characteristics, clinical variables, and hospital occupancy measures, this dynamic vital score was used to train predictive models.ResultsHospitalizations can be predicted with an area under the receiver-operating characteristic curve (AUC) of 92% using symptoms, hospital occupancy, and patient characteristics, including social determinants of health. Parsimonious models to predict intensive care, mechanical ventilation, and mortality that used the most recent labs and vitals exhibited AUCs of 92.7%, 91.2%, and 94%, respectively. Early predictive models, using labs and vital signs closer to admission had AUCs of 81.1%, 84.9%, and 92%, respectively.DiscussionThe most accurate models exhibit racial bias, being more likely to falsely predict that Black patients will be hospitalized. Models that are only based on the dynamic vital score exhibited accuracies close to the best parsimonious models, although the latter also used laboratories.ConclusionsThis large study demonstrates that COVID-19 severity may accurately be predicted using a score that accounts for the dynamic evolution of vital signs. Further, race, social determinants of health, and hospital occupancy play an important role.

Automatic Traffic State Recognition Based on Video Features Extracted by an Autoencoder

Video surveillance has become an important measure of urban traffic monitoring and control. However, due to the complex and diverse video scenes, traffic data extraction from original videos is a sophisticated and difficult task, and corresponding algorithms are of high complexity and calculation cost. To reduce the algorithm complexity and subsequent computation cost, this study proposed an autoencoder model which effectively reduces the video dimension by optimizing structural parameters; thus several traffic recognition models can conduct image processing work based on dimension-reduced videos. Firstly, an optimal autoencoder model A ∗ with five hidden layers was constructed. Then, it was combined with a linear classifier, support vector machine, deep neural network, DNN linear classification method, and the k-means clustering method; thus, five traffic state recognition models were constructed: A ∗ -Linear, A ∗ -SVM, A ∗ -DNN, A ∗ -DNN_Linear, and A ∗ -k-means. Train and test results show that the accuracy rate and recall rate of A ∗ -linear, A ∗ -SVM, A ∗ -DNN, and A ∗ -DNN_Linear were 94.5%–97.1%, and the F1 score was 94.4%–97.1%; besides, the accuracy rate, recall rate, and F1 score of A ∗ -k-means were all approximately 95%, which suggests that the combination of the autoencoder A ∗ and common classification or clustering methods achieve good recognition performance. Comparison was also implemented among the five models proposed above and four CNN-based models such as AlexNet, LeNet, GoogLeNet, and VGG16, which shows that the five proposed modes achieve F1 scores of 94.4%–97.1%, while the four CNN-based models achieve F1 scores of 16.7%–94%, indicating that the proposed light weight design methods outperform more complex CNN-based models in traffic state recognition.

Multivariate Classification of Brain Blood-Oxygen Signal Complexity for the Diagnosis of Children with Tourette Syndrome.

Tourette syndrome (TS) is a childhood-onset neuropsychiatric disorder characterized by the presence of multiple motor and vocal tics. Because of its varied clinical expressions and lack of reliable diagnostic biomarker, present TS diagnosis still depends on qualitative descriptions of symptoms. Our study aimed to investigate whether the complexity of resting state brain activity can serve as a potential biomarker for TS diagnosis, since it has been used successfully in various neuropsychiatric disorders, including two common TS comorbidities: attention-deficit hyperactivity disorder (ADHD) and obsessive-compulsive disorder (OCD). In the current study, we used both univariate analysis and multivariate searchlight analysis with both linear and non-linear classification methods to explore the group differences in the complexity of resting state brainblood oxygen level-dependent (BOLD) signals between 25 TS boys without comorbidity and 25 sex, age and educational years matched healthy controls (HCs). We also investigated the relation between symptom severity in TS patients (YGTSS scores) and complexity indices derived from different analysis methods. We found: i) univariate analysis revealed reduced complexity in TS patients in the left cerebellum, left superior frontal gyrus, and left medial frontal gyrus; ii) multivariate analysis with non-linear classification method achieved thehighest performance (accuracy: 0.94, sensitivity: 0.96, specificity: 0.92, AUC: 0.95) in bilateral supplementary motor areas; iii) significant correlationswere found between complexity index derived from multivariate analysis with non-linear classification method and Tic severity (YGTSS scores) in the left cerebellum (r = 0.523, with YGTSS phonic) and in theright supplementary motor area (r = 0.767, with YGTSS motor). Taken together, theseresults suggested that complexity of resting stateBOLD activity is a highly effective index for differentiating TS patients from normal controls. Ithas agood potential to be a quantitative biomarker for TS diagnosis.

Rapid Differentiation of Unfrozen and Frozen-Thawed Tuna with Non-Destructive Methods and Classification Models: Bioelectrical Impedance Analysis (BIA), Near-Infrared Spectroscopy (NIR) and Time Domain Reflectometry (TDR).

The performances of three non-destructive sensors, based on different principles, bioelectrical impedance analysis (BIA), near-infrared spectroscopy (NIR) and time domain reflectometry (TDR), were studied to discriminate between unfrozen and frozen-thawed fish. Bigeye tuna (Thunnus obesus) was selected as a model to evaluate these technologies. The addition of water and additives is usual in the fish industry, thus, in order to have a wide range of possible commercial conditions, some samples were injected with different water solutions (based on different concentrations of salt, polyphosphates and a protein hydrolysate solution). Three different models, based on partial least squares discriminant analysis (PLS-DA), were developed for each technology. This is a linear classification method that combines the properties of partial least squares (PLS) regression with the classification power of a discriminant technique. The results obtained in the evaluation of the test set were satisfactory for all the sensors, giving NIR the best performance (accuracy = 0.91, error rate = 0.10). Nevertheless, the classification accomplished with BIA and TDR data resulted also satisfactory and almost equally as good, with accuracies of 0.88 and 0.86 and error rates of 0.14 and 0.15, respectively. This work opens new possibilities to discriminate between unfrozen and frozen-thawed fish samples with different non-destructive alternatives, regardless of whether or not they have added water.

APPLICATION OF MACHINE LEARNING ALGORITHMS FOR PROCESSING COMMENTS FROM THE YOUTUBE VIDEO HOSTING UNDER TRAINING VIDEOS

Purpose. Detecting toxic comments on YouTube video hosting under training videos by classifying unstructured text using a combination of machine learning methods. Methodology. To work with the specified type of data, machine learning methods were used for cleaning, normalizing, and presenting textual data in a form acceptable for processing on a computer. Directly to classify comments as “toxic”, we used a logistic regression classifier, a linear support vector classification method without and with a learning method – stochastic gradient descent, a random forest classifier and a gradient enhancement classifier. In order to assess the work of the classifiers, the methods of calculating the matrix of errors, accuracy, completeness and F-measure were used. For a more generalized assessment, a cross-validation method was used. Python programming language. Findings. Based on the assessment indicators, the most optimal methods were selected – support vector machine (Linear SVM), without and with the training method using stochastic gradient descent. The described technologies can be used to analyze the textual comments under any training videos to detect toxic reviews. Also, the approach can be useful for identifying unwanted or even aggressive information on social networks or services where reviews are provided. Originality. It consists in a combination of methods for preprocessing a specific type of text, taking into account such features as the possibility of having a timecode, emoji, links, and the like, as well as in the adaptation of classification methods of machine learning for the analysis of Russian-language comments. Practical value. It is about optimizing (simplification) the comment analysis process. The need for this processing is due to the growing volumes of text data, especially in the field of education through quarantine conditions and the transition to distance learning. The volume of educational Internet content already needs to automate the processing and analysis of feedback, over time this need will only grow.

Otel Oda Fiyatlarını Açıklamada Makine Öğrenmesi Algoritmalarının Kıyaslanması (Comparison of Machine Learning Algorithms in Explaining of Hotel Room Prices)

Purpose: The aim of the study is to analyze some of the main machine learning algorithms using big data obtained by web mining and to test the performance of these algorithms to explain hotel room prices. Thus, it is the determination of the model that best explains the hotel room prices. Design/methodology/approach – Web mining/scraping method was used to obtain research data. The target website was scanned for about six months with the help of an algorithm, and the data obtained from 6558 accommodation facilities were used in analysis. The second part of the research consists of statistical analysis and comparison of machine learning algorithms. Python programming language was used for analysis and implementation of algorithms. Pandas, numpy libraries for data processing; seaborn, matplotlib for graphics and visualization; scikit-learn is used to run machine learning algorithms. After the analysis, a model was created by logistic regression, which was thought to be the most suitable for the data. Results: It is seen that the compared Random Forest and Decision Tree algorithms both explain the data set at a rate of approximately 99.89%, so the tree/branching has been successful. The KNN algorithm achieved the highest performance with a classification of three clusters at 62.12%. Logistic Regression, Stochastic Gradient Decent and Support Vector Machines using the linear classification method obtained the highest score with 39.13% logistic regression method. In the model created by logistic regression, the score given to the hotel by the guests, the rank of the hotel among other hotels in the region, the type of the hotel and the city in which it is located were found to be statistically significant (p <0.05). Discussion: As a result of the research, machine learning algorithms were compared using the hotel room price data that obtained from web mining/scraping. It was also found important information about the hotel room rates in Turkey. A regression model has revealed which of the 44 independent variables are significant in explaining the hotel room price.

A novel linear assorted classification method based association rule mining with spatial data

A novel linear assorted classification method based association rule mining with spatial data

Missing Value Imputation in Stature Estimation by Learning Algorithms Using Anthropometric Data: A Comparative Study

Estimating stature is essential in the process of personal identification. Because it is difficult to find human remains intact at crime scenes and disaster sites, for instance, methods are needed for estimating stature based on different body parts. For instance, the upper and lower limbs may vary depending on ancestry and sex, and it is of great importance to design adequate methodology for incorporating these in estimating stature. In addition, it is necessary to use machine learning rather than simple linear regression to improve the accuracy of stature estimation. In this study, the accuracy of statures estimated based on anthropometric data was compared using three imputation methods. In addition, by comparing the accuracy among linear and nonlinear classification methods, the best method was derived for estimating stature based on anthropometric data. For both sexes, multiple imputation was superior when the missing data ratio was low, and mean imputation performed well when the ratio was high. The support vector machine recorded the highest accuracy in all ratios of missing data. The findings of this study showed appropriate imputation methods for estimating stature with missing anthropometric data. In particular, the machine learning algorithms can be effectively used for estimating stature in humans.

RETRACTED ARTICLE: Normal discriminant deep convolution neural classification based web behavioral pattern mining for user identification

Mining the user behavior patterns from huge log files plays a major role in discovering the web user identity. Several data mining techniques describe the way of predicting the user identity using web access behavior patterns. Accurate prediction of the user identity is still a challenging issue. The proposed work introduces a normal discriminant Tanimoto similarity based deep convolution feedforward neural learning classification (NDTS-DCFNLC) technique to improve the user identifications with their web access behavior patterns. The NDTS-DCFNLC technique comprises multiple layers. Normal discriminant preprocessing is carried out in layer one to remove the unwanted patterns from the web access log files. Similarity among the relevant web pattern is calculated using Tanimoto similarity at layer two. Sigmoid activation function is used in output layer to classify the frequently accessed patterns with higher accuracy and minimum error based on similarity threshold. Experimental testing is done using apache weblog with different number of patterns and the diverse constraints such as classification accuracy, false positive rate, space requirements and execution time. Results were compared with linear temporal logic and KNN classification methods. The outcome shows NDTS-DCFNLC technique increases classification accuracy, decreases the false positives, execution time and the space requirements.