Recognition of Toxicity of Reviews in Online Discussions

Machová, Kristína; Mach, Marián; Vasilko, Matej

doi:10.12700/aph.19.4.2022.4.1

Cited by 6 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Bagging is a decision tree-based ensemble method that generates multiple resampled training data by sampling subjects with replacement from training data, creates a decision tree from each resampled training data, and classi es the groups by combining decision trees through majority votes (Machová et al, 2006). The tuning parameters used in this model are the minimum number of subjects included in the nal node ("minbucket") and the maximum number of times allowed to overlap input variables used when creating decision trees ("maxdepth").…”

Section: Baggingmentioning

confidence: 99%

Identification of Biomarkers in Gynecologic Cancers: A Machine Learning Approach for Metabolomics

Lee,

Cha,

Lee

et al. 2024

Preprint

View full text Add to dashboard Cite

Introduction Diagnostic methods for gynecologic cancer (GC) such as cervical cancer (CC), endometrial cancer (EC), and ovarian cancer (OC) remain poorly developed. Machine learning (ML) algorithms have recently been compared to traditional statistical methods utilized to analyze metabolomics data. Objective This study aimed to identify the clinical metabolic markers associated with GCs by comparing ML algorithms with orthogonal partial least squares-discriminant analysis (OPLS-DA). Methods Untargeted metabolomic analysis was performed on plasma from 42 patients with GC (24 CC, 9 EC, and 9 OC) and 57 healthy female participants. GC and healthy control groups were classified using OPLS-DA and eight ML algorithms. The ML algorithm with the best classification performance was used to assess CC, EC, and OC with healthy controls, and metabolite candidates involved in each GC were selected. Results Upon comparing the classification model performance between the GC and control groups, random forest (RF) model displayed the best performance with an area under the curve (AUC) of 0.9999. The multi-classification RF model was established to distinguish all four groups and was achieved an AUC of 0.8351. The AUCs of the three GC subgroup assessment RF models comparing patients with CC, EC, and OC with healthy controls were 0.9838, 0.7500, and 0.7321, respectively. Plasma concentrations of two identified metabolites significantly increased in patients with GCs. Conclusion Several ML algorithms were used to distinguish GC, showed better performance than conventional OPLS-DA. Proline betaine and lysophosphatidyl ethanolamine (18:0/0:0) selected in RF models were suggested as metabolite candidates associated with GCs.

show abstract

Section: Baggingmentioning

confidence: 99%

Identification of Biomarkers in Gynecologic Cancers: A Machine Learning Approach for Metabolomics

Lee,

Cha,

Lee

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Α represents the weight and 𝐻 𝑚 (𝑑𝑖) = For the instance di, this is the prediction of the m th classifier [30].…”

Section: Bagging Treementioning

confidence: 99%

Machine Learning Approach to Predict Cardiovascular Disease in Bangladesh: Evidence from a Cross-Sectional Study in 2023.

Hossain,

Hasan,

Faruk

et al. 2023

Preprint

View full text Add to dashboard Cite

Background Cardiovascular disorders (CVDs) are widely considered the leading cause of death worldwide. Lower and middle-income countries (LMICs) like Bangladesh are also affected by several types of CVDs such as heart failure and stroke. The leading factors of death in Bangladesh have increasingly switched from severe infections and parasitic illness to CVDs recently. Materials and methods The study dataset is a random sample of the 391 CVD patients' medical records collected between August 2022 and April 2023 using simple random sampling. Moreover, 260 data are also collected from individuals with no CVD problem for comparison purposes. Crosstabs and chi-square are used to find the association between CVD and explanatory variables. Logistic regression, Naïve Bayes classifier, Decision Tree, AdaBoost classifier, Random Forest, Bagging Tree, and Ensemble learning classifiers are used to predict CVD in this study. The performance evaluations encompassed accuracy, sensitivity, specificity, and the area under the receiver operator characteristic (AU-ROC) curve. Result Random Forest has the highest precision among the five techniques considered. The precision rates for the mentioned classifiers are as follows: Logistic Regression (93.67%), Naïve Bayes (94.87%), Decision Tree (96.1%), AdaBoost (94.94%), Random Forest (96.15%), and Bagging Tree (94.87%). The Random Forest classifier maintains the highest balance between correct and incorrect predictions. With 98.04% accuracy, the Random Forest Classifier achieves the best precision (96.15%), robust recall (100%), and a high F1 score (97.7%). In contrast, the Logistic Regression model achieves the lowest accuracy at 95.42%. Remarkably, the Random Forest classifier attains the highest AUC value (0.989). Conclusion This research is mainly focused on identifying factors that are critical in impacting CVD patients and predicting CVD risk. It is strongly advised that the Random Forest technique be implemented in the system for predicting cardiac disease. This research may change clinical practice by giving doctors a new instrument to determine a patient's prognosis for CVD.

show abstract

“…As an instance, the random forest approach incorporates random decision trees along with bagging to acquire extremely elevated classifcation precision. Bagging attempts to execute parallel trainees on undersized sample inhabitants and then carries a norm of all the forecasts [48]. Bagging operates by integrating forecasts by voting, every model obtains equivalent signifcance "Idealized" interpretation: Model several training groups of size n and then create a classifer for each training group and connect the classifers' forecasts [49].…”

Section: Bagging Classifermentioning

confidence: 99%

Cognitive Lightweight Logistic Regression-Based IDS for IoT-Enabled FANET to Detect Cyberattacks

Rahman

Aziz

Usman

et al. 2023

Mobile Information Systems

View full text Add to dashboard Cite

In recent few years, flying ad hoc networks are utilized more for interconnectivity. In the topological scenario of FANETs, IoT nodes are available on ground where UAVs collect information. Due to high mobility patterns of UAVs cause disruption where intruders easily deploy cyberattacks like DoS/DDoS. Flying ad hoc networks use to have UAVs, satellite, and base station in the physical structure. IoT-based UAV networks are having many applications which include agriculture, rescue operations, tracking, and surveillance. However, DoS/DDoS attacks disturb the behaviour of entire FANET which lead to unbalance energy, end-to-end delay, and packet loss. This research study is focused about the detail study of machine learning-based IDS. Also, cognitive lightweight-LR approach is modeled using UNSW-NB 15 dataset. IoT-based UAV network is introduced using machine learning to detect possible security attacks. The queuing and data traffic model is utilized to implement DT, RF, XGBoost, AdaBoost, Bagging and logistic regression in the environment of IoT-based UAV network. Logistic regression is the proposed approach which is used to estimate statistical possibility. Overall, experimentation is based on binomial distribution. There exists linear association approach in logistic regression. In comparison with other techniques, logistic regression behaviour is lightweight and low cost. The simulation results presents logistic regression better results in contrast with other techniques. Also, high accuracy is balanced well in optimal way.

show abstract

Recognition of Toxicity of Reviews in Online Discussions

Cited by 6 publications

References 0 publications

Identification of Biomarkers in Gynecologic Cancers: A Machine Learning Approach for Metabolomics

Identification of Biomarkers in Gynecologic Cancers: A Machine Learning Approach for Metabolomics

Machine Learning Approach to Predict Cardiovascular Disease in Bangladesh: Evidence from a Cross-Sectional Study in 2023.

Cognitive Lightweight Logistic Regression-Based IDS for IoT-Enabled FANET to Detect Cyberattacks

Contact Info

Product

Resources

About