ANALYSIS OF THE INFLUENCE OF MACHINE LEARNING ALGORITHM PARAMETERS ON THE RESULTS OF TRAFFIC CLASSIFICATION IN REAL TIME

The paper analyzes the impact of setting the parameters of Machine Learning algorithms on the results of traffic classification in real-time. The Random Forest and XGBoost algorithms are considered. A brief description of the work of both methods and methods for evaluating the results of classification is given. Experimental studies are conducted on a database obtained on a real network, separately for TCP and UDP flows. In order for the results of the study to be used in real time, a special feature matrix is created based on the first 15 packets of the flow. The main parameters of the Random Forest (RF) algorithm for configuration are the number of trees, the partition criterion used, the maximum number of features for constructing the partition function, the depth of the tree, and the minimum number of samples in the node and in the leaf. For XGBoost, the number of trees, the depth of the tree, the minimum number of samples in the leaf, for features, and the percentage of samples needed to build the tree are taken. Increasing the number of trees leads to an increase in accuracy to a certain value, but as shown in the article, it is important to make sure that the model is not overfitted. To combat overfitting, the remaining parameters of the trees are used. In the data set under study, by eliminating overfitting, it was possible to achieve an increase in classification accuracy for individual applications by 11-12% for Random Forest and by 12-19% for XGBoost. The results show that setting the parameters is a very important step in building a traffic classification model, because it helps to combat overfitting and significantly increases the accuracy of the algorithm’s predictions. In addition, it was shown that if the parameters are properly configured, XGBoost, which is not very popular in traffic classification works, becomes a competitive algorithm and shows better results compared to the widespread Random Forest.

Download Full-text

Application of random forest regression to the calculation of gas-phase chemistry within the GEOS-Chem chemistry model v10

Geoscientific Model Development ◽

10.5194/gmd-12-1209-2019 ◽

2019 ◽

Vol 12 (3) ◽

pp. 1209-1225 ◽

Cited By ~ 15

Author(s):

Christoph A. Keller ◽

Mat J. Evans

Keyword(s):

Machine Learning ◽

Random Forest ◽

Gas Phase ◽

Atmospheric Chemistry ◽

Random Forest Regression ◽

Data Set ◽

Gas Phase Chemistry ◽

Chemical Conditions ◽

Phase Chemistry ◽

The Impact

Abstract. Atmospheric chemistry models are a central tool to study the impact of chemical constituents on the environment, vegetation and human health. These models are numerically intense, and previous attempts to reduce the numerical cost of chemistry solvers have not delivered transformative change. We show here the potential of a machine learning (in this case random forest regression) replacement for the gas-phase chemistry in atmospheric chemistry transport models. Our training data consist of 1 month (July 2013) of output of chemical conditions together with the model physical state, produced from the GEOS-Chem chemistry model v10. From this data set we train random forest regression models to predict the concentration of each transported species after the integrator, based on the physical and chemical conditions before the integrator. The choice of prediction type has a strong impact on the skill of the regression model. We find best results from predicting the change in concentration for long-lived species and the absolute concentration for short-lived species. We also find improvements from a simple implementation of chemical families (NOx = NO + NO2). We then implement the trained random forest predictors back into GEOS-Chem to replace the numerical integrator. The machine-learning-driven GEOS-Chem model compares well to the standard simulation. For ozone (O3), errors from using the random forests (compared to the reference simulation) grow slowly and after 5 days the normalized mean bias (NMB), root mean square error (RMSE) and R2 are 4.2 %, 35 % and 0.9, respectively; after 30 days the errors increase to 13 %, 67 % and 0.75, respectively. The biases become largest in remote areas such as the tropical Pacific where errors in the chemistry can accumulate with little balancing influence from emissions or deposition. Over polluted regions the model error is less than 10 % and has significant fidelity in following the time series of the full model. Modelled NOx shows similar features, with the most significant errors occurring in remote locations far from recent emissions. For other species such as inorganic bromine species and short-lived nitrogen species, errors become large, with NMB, RMSE and R2 reaching >2100 % >400 % and <0.1, respectively. This proof-of-concept implementation takes 1.8 times more time than the direct integration of the differential equations, but optimization and software engineering should allow substantial increases in speed. We discuss potential improvements in the implementation, some of its advantages from both a software and hardware perspective, its limitations, and its applicability to operational air quality activities.

Download Full-text

MONITORING AND CLASSIFYING THE STATE OF HARD DISKS USING RECURRENT NEURAL NETWORKS

Kontrol Diagnostika ◽

10.14489/td.2021.10.pp.036-043 ◽

2021 ◽

pp. 36-43

Author(s):

L. A. Demidova ◽

A. V. Filatov

Keyword(s):

Neural Networks ◽

Recurrent Neural Networks ◽

Experimental Studies ◽

Machine Learning Algorithms ◽

Classification Model ◽

Dense Layer ◽

Hard Disks ◽

Destructive Testing ◽

Data Set ◽

State Classification

The article considers an approach to solving the problem of monitoring and classifying the states of hard disks, which is solved on a regular basis, within the framework of the concept of non-destructive testing. It is proposed to solve this problem by developing a classification model using machine learning algorithms, in particular, using recurrent neural networks with Simple RNN, LSTM and GRU architectures. To develop a classification model, a data set based on the values of SMART sensors installed on hard disks it used. It represents a group of multidimensional time series. At the same time, the structure of the classification model contains two layers of a neural network with one of the recurrent architectures, as well as a Dropout layer and a Dense layer. The results of experimental studies confirming the advantages of LSTM and GRU architectures as part of hard disk state classification models are presented.

Download Full-text

False Positive RFID Detection Using Classification Models

Applied Sciences ◽

10.3390/app9061154 ◽

2019 ◽

Vol 9 (6) ◽

pp. 1154 ◽

Cited By ~ 11

Author(s):

Ganjar Alfian ◽

Muhammad Syafrudin ◽

Bohan Yoon ◽

Jongtae Rhee

Keyword(s):

Machine Learning ◽

Supply Chain ◽

Real Time ◽

Outlier Detection ◽

Radio Frequency Identification ◽

False Positives ◽

Machine Learning Algorithms ◽

Classification Model ◽

Automated Identification ◽

Rfid Data

Radio frequency identification (RFID) is an automated identification technology that can be utilized to monitor product movements within a supply chain in real-time. However, one problem that occurs during RFID data capturing is false positives (i.e., tags that are accidentally detected by the reader but not of interest to the business process). This paper investigates using machine learning algorithms to filter false positives. Raw RFID data were collected based on various tagged product movements, and statistical features were extracted from the received signal strength derived from the raw RFID data. Abnormal RFID data or outliers may arise in real cases. Therefore, we utilized outlier detection models to remove outlier data. The experiment results showed that machine learning-based models successfully classified RFID readings with high accuracy, and integrating outlier detection with machine learning models improved classification accuracy. We demonstrated the proposed classification model could be applied to real-time monitoring, ensuring false positives were filtered and hence not stored in the database. The proposed model is expected to improve warehouse management systems by monitoring delivered products to other supply chain partners.

Download Full-text

Spoken words as biomarkers: using machine learning to gain insight into communication as a predictor of anxiety

Journal of the American Medical Informatics Association ◽

10.1093/jamia/ocaa049 ◽

2020 ◽

Vol 27 (6) ◽

pp. 929-933

Author(s):

George Demiris ◽

Kristin L Corey Magan ◽

Debra Parker Oliver ◽

Karla T Washington ◽

Chad Chadwick ◽

...

Keyword(s):

Machine Learning ◽

Secondary Data ◽

Health Indicators ◽

Machine Learning Algorithms ◽

Standardized Assessments ◽

Learning Tools ◽

Data Set ◽

Problem Solving Therapy ◽

Audio Communication ◽

The Impact

Abstract Objective The goal of this study was to explore whether features of recorded and transcribed audio communication data extracted by machine learning algorithms can be used to train a classifier for anxiety. Materials and Methods We used a secondary data set generated by a clinical trial examining problem-solving therapy for hospice caregivers consisting of 140 transcripts of multiple, sequential conversations between an interviewer and a family caregiver along with standardized assessments of anxiety prior to each session; 98 of these transcripts (70%) served as the training set, holding the remaining 30% of the data for evaluation. Results A classifier for anxiety was developed relying on language-based features. An 86% precision, 78% recall, 81% accuracy, and 84% specificity were achieved with the use of the trained classifiers. High anxiety inflections were found among recently bereaved caregivers and were usually connected to issues related to transitioning out of the caregiving role. This analysis highlighted the impact of lowering anxiety by increasing reciprocity between interviewers and caregivers. Conclusion Verbal communication can provide a platform for machine learning tools to highlight and predict behavioral health indicators and trends.

Download Full-text

Distinguishing Focal Cortical Dysplasia From Glioneuronal Tumors in Patients With Epilepsy by Machine Learning

Frontiers in Neurology ◽

10.3389/fneur.2020.548305 ◽

2020 ◽

Vol 11 ◽

Author(s):

Yi Guo ◽

Yushan Liu ◽

Wenjie Ming ◽

Zhongjin Wang ◽

Junming Zhu ◽

...

Keyword(s):

Machine Learning ◽

Random Forest ◽

Focal Cortical Dysplasia ◽

Cortical Dysplasia ◽

Machine Learning Algorithms ◽

Classification Model ◽

Supervised Machine Learning ◽

Seizure Onset ◽

Glioneuronal Tumors ◽

Patients With Epilepsy

Purpose: We are aiming to build a supervised machine learning-based classifier, in order to preoperatively distinguish focal cortical dysplasia (FCD) from glioneuronal tumors (GNTs) in patients with epilepsy.Methods: This retrospective study was comprised of 96 patients who underwent epilepsy surgery, with the final neuropathologic diagnosis of either an FCD or GNTs. Seven classical machine learning algorithms (i.e., Random Forest, SVM, Decision Tree, Logistic Regression, XGBoost, LightGBM, and CatBoost) were employed and trained by our dataset to get the classification model. Ten features [i.e., Gender, Past history, Age at seizure onset, Course of disease, Seizure type, Seizure frequency, Scalp EEG biomarkers, MRI features, Lesion location, Number of antiepileptic drug (AEDs)] were analyzed in our study.Results: We enrolled 56 patients with FCD and 40 patients with GNTs, which included 29 with gangliogliomas (GGs) and 11 with dysembryoplasic neuroepithelial tumors (DNTs). Our study demonstrated that the Random Forest-based machine learning model offered the best predictive performance on distinguishing the diagnosis of FCD from GNTs, with an F1-score of 0.9180 and AUC value of 0.9340. Furthermore, the most discriminative factor between FCD and GNTs was the feature “age at seizure onset” with the Chi-square value of 1,213.0, suggesting that patients who had a younger age at seizure onset were more likely to be diagnosed as FCD.Conclusion: The Random Forest-based machine learning classifier can accurately differentiate FCD from GNTs in patients with epilepsy before surgery. This might lead to improved clinician confidence in appropriate surgical planning and treatment outcomes.

Download Full-text

Machine learning algorithms can predict tail biting outbreaks in pigs using feeding behaviour records

10.1101/2021.05.11.443554 ◽

2021 ◽

Author(s):

Catherine Ollagnier ◽

Claudia Kasper ◽

Anna Wallenbeck ◽

Linda Keeling ◽

Siavash A Bigdeli

Keyword(s):

Machine Learning ◽

Random Forest ◽

Real Time ◽

Feeding Behaviour ◽

Learning Algorithm ◽

Learning Algorithms ◽

Machine Learning Algorithms ◽

Machine Learning Algorithm ◽

Tail Biting ◽

Testing Set

Tail biting is a detrimental behaviour that impacts the welfare and health of pigs. Early detection of tail biting precursor signs allows for preventive measures to be taken, thus avoiding the occurrence of the tail biting event. This study aimed to build a machine-learning algorithm for real time detection of upcoming tail biting outbreaks, using feeding behaviour data recorded by an electronic feeder. Prediction capacities of seven machine learning algorithms (e.g., random forest, neural networks) were evaluated from daily feeding data collected from 65 pens originating from 2 herds of grower-finisher pigs (25-100kg), in which 27 tail biting events occurred. Data were divided into training and testing data, either by randomly splitting data into 75% (training set) and 25% (testing set), or by randomly selecting pens to constitute the testing set. The random forest algorithm was able to predict 70% of the upcoming events with an accuracy of 94%, when predicting events in pens for which it had previous data. The detection of events for unknown pens was less sensitive, and the neural network model was able to detect 14% of the upcoming events with an accuracy of 63%. A machine-learning algorithm based on ongoing data collection should be considered for implementation into automatic feeder systems for real time prediction of tail biting events.

Download Full-text

Breast Cancer Detection Using Machine Learning Algorithms

International Journal of Computer Science and Mobile Computing ◽

10.47760/ijcsmc.2021.v10i11.002 ◽

2021 ◽

Vol 10 (11) ◽

pp. 4-11

Author(s):

Samer Hamed ◽

Abdelwadood Mesleh ◽

Abdullah Arabiyyat

Keyword(s):

Machine Learning ◽

Random Forest ◽

Cancer Detection ◽

Computer Aided Design ◽

Machine Learning Algorithms ◽

Breast Cancers ◽

Data Set ◽

Cad System ◽

Aided Design ◽

F Measure

This paper presents a computer-aided design (CAD) system that detects breast cancers (BCs). BC detection uses random forest, AdaBoost, logistic regression, decision trees, naïve Bayes and conventional neural networks (CNNs) classifiers, these machine learning (ML) based algorithms are trained to predicting BCs (malignant or benign) on BC Wisconsin data-set from the UCI repository, in which attribute clump thickness is used as evaluation class. The effectiveness of these ML algorithms are evaluated in terms of accuracy and F-measure; random forest outperformed the other classifiers and achieved 99% accuracy and 99% F-measure.

Download Full-text

A Self-Supervised Machine Learning Approach for Objective Live Cell Segmentation and Analysis

10.21203/rs.3.rs-147010/v1 ◽

2021 ◽

Author(s):

Marc Raphael ◽

Michael Robitaille ◽

Jeff Byers ◽

Joseph Christodoulides

Keyword(s):

Machine Learning ◽

Learning Algorithm ◽

Learning Algorithms ◽

Label Cell ◽

Live Cell ◽

Machine Learning Algorithms ◽

Classification Model ◽

Supervised Machine Learning ◽

Cell Segmentation ◽

Data Set

Abstract Machine learning algorithms hold the promise of greatly improving live cell image analysis by way of (1) analyzing far more imagery than can be achieved by more traditional manual approaches and (2) by eliminating the subjective nature of researchers and diagnosticians selecting the cells or cell features to be included in the analyzed data set. Currently, however, even the most sophisticated model based or machine learning algorithms require user supervision, meaning the subjectivity problem is not removed but rather incorporated into the algorithm’s initial training steps and then repeatedly applied to the imagery. To address this roadblock, we have developed a self-supervised machine learning algorithm that recursively trains itself directly from the live cell imagery data, thus providing objective segmentation and quantification. The approach incorporates an optical flow algorithm component to self-label cell and background pixels for training, followed by the extraction of additional feature vectors for the automated generation of a cell/background classification model. Because it is self-trained, the software has no user-adjustable parameters and does not require curated training imagery. The algorithm was applied to automatically segment cells from their background for a variety of cell types and five commonly used imaging modalities - fluorescence, phase contrast, differential interference contrast (DIC), transmitted light and interference reflection microscopy (IRM). The approach is broadly applicable in that it enables completely automated cell segmentation for long-term live cell phenotyping applications, regardless of the input imagery’s optical modality, magnification or cell type.

Download Full-text

A Self-Supervised Machine Learning Approach for Objective Live Cell Segmentation and Analysis

10.1101/2021.01.07.425773 ◽

2021 ◽

Author(s):

Michael C. Robitaille ◽

Jeff M. Byers ◽

Joseph A. Christodoulides ◽

Marc P. Raphael

Keyword(s):

Machine Learning ◽

Learning Algorithm ◽

Learning Algorithms ◽

Label Cell ◽

Live Cell ◽

Machine Learning Algorithms ◽

Classification Model ◽

Supervised Machine Learning ◽

Cell Segmentation ◽

Data Set

Machine learning algorithms hold the promise of greatly improving live cell image analysis by way of (1) analyzing far more imagery than can be achieved by more traditional manual approaches and (2) by eliminating the subjective nature of researchers and diagnosticians selecting the cells or cell features to be included in the analyzed data set. Currently, however, even the most sophisticated model based or machine learning algorithms require user supervision, meaning the subjectivity problem is not removed but rather incorporated into the algorithm's initial training steps and then repeatedly applied to the imagery. To address this roadblock, we have developed a self-supervised machine learning algorithm that recursively trains itself directly from the live cell imagery data, thus providing objective segmentation and quantification. The approach incorporates an optical flow algorithm component to self-label cell and background pixels for training, followed by the extraction of additional feature vectors for the automated generation of a cell/background classification model. Because it is self-trained, the software has no user-adjustable parameters and does not require curated training imagery. The algorithm was applied to automatically segment cells from their background for a variety of cell types and five commonly used imaging modalities - fluorescence, phase contrast, differential interference contrast (DIC), transmitted light and interference reflection microscopy (IRM). The approach is broadly applicable in that it enables completely automated cell segmentation for long-term live cell phenotyping applications, regardless of the input imagery's optical modality, magnification or cell type.

Download Full-text

Fast and Effective Techniques for LWIR Radiative Transfer Modeling: A Dimension Reduction Approach

10.20944/preprints201906.0062.v1 ◽

2019 ◽

Author(s):

Nicholas Westing ◽

Brett Borghetti ◽

Kevin Gross

Keyword(s):

Machine Learning ◽

Radiative Transfer ◽

Dimension Reduction ◽

Real Time ◽

Computational Cost ◽

Hyperspectral Data ◽

Machine Learning Algorithms ◽

Atmospheric Effects ◽

Atmospheric Compensation ◽

The Impact

The increasing spatial and spectral resolution of hyperspectral imagers yields detailed spectroscopy measurements from both space-based and airborne platforms. Machine learning algorithms have achieved state-of-the-art material classification performance on benchmark hyperspectral data sets; however, these techniques often do not consider varying atmospheric conditions experienced in a real-world detection scenario. To reduce the impact of atmospheric effects in the at-sensor signal, atmospheric compensation must be performed. Radiative Transfer (RT) modeling can generate high-fidelity atmospheric estimates at detailed spectral resolutions, but is often too time-consuming for real-time detection scenarios. This research utilizes machine learning methods to perform dimension reduction on the transmittance, upwelling radiance, and downwelling radiance (TUD) data to create high accuracy atmospheric estimates with lower computational cost than RT modeling. The utility of this approach is investigated using the instrument line shape for the Mako long-wave infrared hyperspectral sensor. This study employs physics-based metrics and loss functions to identify promising dimension reduction techniques. As a result, TUD vectors can be produced in real-time allowing for atmospheric compensation across diverse remote sensing scenarios.

Download Full-text