pith. machine review for the scientific record. sign in

arxiv: 2605.11997 · v2 · submitted 2026-05-12 · 💻 cs.CR

Recognition: no theorem link

A microservices-based endpoint monitoring platform with predictive NLP models for real-time security and hate-speech risk alerting

Anubis Graciela De Moraes Rossetto, Darlan Noetzold, Juan Francisco De Paz Santana, Valderi Reis Quietinho Leithard

Pith reviewed 2026-05-14 20:49 UTC · model grok-4.3

classification 💻 cs.CR
keywords microservicesendpoint monitoringhate-speech detectionNLP modelssecurity alertingBERTreal-time alertsdata exfiltration
0
0 comments X

The pith

A microservices platform collects endpoint telemetry and applies NLP models to deliver real-time alerts on security risks and hate speech.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper describes a unified platform that gathers telemetry from endpoint devices and applies natural language processing to detect security threats and hateful content in real time. Built on microservices, it routes events with RabbitMQ and uses Redis for quick access, allowing alerts to be centralized instead of handled by separate tools. Tests show BERT models reach 87% accuracy for identifying hate-speech risks, and the setup surfaces signs of data leaks and rule breaches promptly. This integrated approach addresses the problem of isolated solutions that delay responses to incidents in corporate settings.

Core claim

The authors claim that their microservices-based endpoint monitoring platform, incorporating predictive NLP models, can promptly surface indicators of data exfiltration and policy violations while centralizing alert management in an integrated framework that combines monitoring, security analytics, and predictive capabilities, with transformer models like BERT achieving an average accuracy of 87% for hate-speech risk detection.

What carries the argument

A modular microservices architecture relying on RabbitMQ for event ingestion and routing and Redis for low-latency data access and alert delivery, along with BERT-based models for text classification.

If this is right

  • The platform can promptly surface indicators of data exfiltration and policy violations.
  • It centralizes alert management across monitoring, security analytics, and predictive capabilities.
  • It provides an integrated framework combining these elements for faster incident response.
  • Transformer models such as BERT achieve an average accuracy of 87% on hate-speech risk detection.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Such a platform could enable faster correlation of user behavior signals across devices and communications.
  • It may support proactive compliance by flagging risks before they escalate in workplace environments.
  • Extending the system to additional risk categories or languages could increase its utility for global organizations.
  • Real-world performance would benefit from calibration on diverse corporate data streams beyond the initial tests.

Load-bearing premise

The 87% accuracy reported for hate-speech detection using BERT will maintain low false positive rates and support effective real-time alerting when applied to actual ongoing corporate endpoint data without further adjustments.

What would settle it

Running the platform on live endpoint streams and comparing its alerts against ground-truth incidents identified by security teams to check for missed detections or excessive false alarms.

Figures

Figures reproduced from arXiv: 2605.11997 by Anubis Graciela De Moraes Rossetto, Darlan Noetzold, Juan Francisco De Paz Santana, Valderi Reis Quietinho Leithard.

Figure 1
Figure 1. Figure 1: Detailed architecture and data flows of the proposed framework. Employee Endpoint (Monitoring Agent) Parallel collectors (sniffer / key capture / etc.) Alert Evidence (metadata + context) Central API Layer (ingestion & orchestration) Auth & User Service Policy Service Alert Service Prediction API Front-End Application Relational Database Evidence Storage Cache Layer Async Messaging Observability Endpoint L… view at source ↗
Figure 2
Figure 2. Figure 2: Alert generation flow. • Scanner: Performs scans to find open ports on a specific target. It runs in a separate Thread and uses the Socket library to establish connections and collect service banners on open ports. Ports and banners are stored in lists, and banners are compared with a dynamically updated list of known vulnerabilities. This list is obtained from public vulnerability databases, such as the N… view at source ↗
Figure 3
Figure 3. Figure 3: Test execution on JMeter You can then analyze [PITH_FULL_IMAGE:figures/full_fig_p021_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: illustrate the trends for response time, CPU usage, and memory heap usage as the number of parallel requests increases. Response time shows an upward trend, represented by the regression formula 74.781 + 0.082x, which estimates response time based on request volume, enabling predictions of the application’s behavior under heavier loads. Similarly, the CPU and memory usage graphs reveal a direct relationshi… view at source ↗
Figure 5
Figure 5. Figure 5: CPU and Memory usage of the Spyware component during a 10-minute execution period. In conclusion, the data capture and processing stages were implemented with a 26 [PITH_FULL_IMAGE:figures/full_fig_p026_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Number of ’fold’ versus Accuracy Figures 7 and 8 show the results of the Balanced Accuracy and the AUC-ROC Curve, respectively, which indicate the effectiveness of predicting hate speech using different strategies. The results for the three models are quite similar, ranging from 75% to 90% for the different fold configurations [PITH_FULL_IMAGE:figures/full_fig_p027_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Number of ’fold’ versus Balanced Accuracy It was noted that each of the models performed similarly and acceptably, maintain￾ing an average accuracy of 87%. However, the Multinomial Naive Bayes model showed a slight advantage over the others. This difference in performance may be related to 27 [PITH_FULL_IMAGE:figures/full_fig_p027_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Number of ’fold’ versus Area Under the ROC Curve the specific nature of the dataset and the suitability of the algorithm for the problem in question. The convergence behavior of these models was consistent, as demonstrated by the use of robust optimization techniques, such as the ”saga” solver for Logistic Regression and the fine-tuning of C and γ parameters in the Support Vector Machine. These settings en… view at source ↗
Figure 9
Figure 9. Figure 9: illustrates the evaluation of the classification models using the Polygon Area Metric (PAM), which combines multiple performance indicators into a single geometric visualization (Aydemir, 2020). Instead of analyzing each metric independently, PAM represents the model as a polygon whose vertices correspond to normalized metric values. In this study, the polygon axes correspond to AUC-ROC, Precision, Recall,… view at source ↗
read the original abstract

Organizations increasingly depend on endpoint devices and corporate communication channels, yet they still face critical risks such as sensitive data leakage, suspicious user behavior, and the circulation of hateful or harmful language in workplace contexts. Current solutions frequently address these issues in isolation (e.g., productivity tracking, data loss prevention, or hate-speech detection), limiting correlation across signals and delaying incident response. This work proposes a unified, microservices-based platform that collects endpoint telemetry and applies predictive natural language processing models to support real-time security and compliance alerting. The architecture is modular and scalable, relying on RabbitMQ for event ingestion and routing and Redis for low-latency data access and alert delivery. For text classification, transformer-based models such as BERT are evaluated for hate-speech risk detection, achieving an average accuracy of 87\%. Experimental results indicate that the proposed platform can promptly surface indicators of data exfiltration and policy violations while centralizing alert management, providing an integrated framework that combines monitoring, security analytics, and predictive capabilities.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 0 minor

Summary. The manuscript proposes a microservices-based endpoint monitoring platform that collects telemetry from devices, routes events via RabbitMQ, caches data with Redis, and applies transformer models such as BERT for hate-speech risk detection (reported 87% average accuracy) to enable real-time security and compliance alerting for issues including data exfiltration and policy violations.

Significance. If the accuracy and real-time performance claims are substantiated, the work would provide a practical integrated framework that correlates endpoint signals with predictive content analysis, potentially improving incident response times in corporate settings over siloed tools.

major comments (1)
  1. [Experimental Results (referenced in abstract)] The abstract and experimental results claim an average accuracy of 87% for BERT-based hate-speech detection and prompt surfacing of data exfiltration indicators, but provide no dataset description, label distribution, training/validation splits, fine-tuning procedure, baseline comparisons, error bars, or latency/throughput measurements for the end-to-end pipeline. This absence makes it impossible to assess overfitting, domain shift to corporate streams, or false-positive rates that would support the alerting use case.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive and detailed review. We agree that the experimental methodology requires substantial elaboration to support the reported accuracy and real-time claims. We address the major comment below and will revise the manuscript accordingly.

read point-by-point responses

  1. Referee: The abstract and experimental results claim an average accuracy of 87% for BERT-based hate-speech detection and prompt surfacing of data exfiltration indicators, but provide no dataset description, label distribution, training/validation splits, fine-tuning procedure, baseline comparisons, error bars, or latency/throughput measurements for the end-to-end pipeline. This absence makes it impossible to assess overfitting, domain shift to corporate streams, or false-positive rates that would support the alerting use case.

    Authors: We acknowledge that the current manuscript version omits critical experimental details, making independent assessment of the 87% accuracy claim difficult. In the revised manuscript we will expand the Experimental Results section with: (1) full dataset descriptions, including sources (public corpora such as the Davidson hate-speech dataset and any internal corporate logs), ethical data-handling notes, label distributions, and exact train/validation/test splits; (2) the BERT fine-tuning procedure, including hyperparameters, optimizer, epochs, and regularization; (3) baseline comparisons (e.g., TF-IDF+SVM, LSTM, RoBERTa) with accuracy, precision, recall, and F1 scores; (4) error bars or standard deviations from repeated runs or cross-validation; and (5) end-to-end latency and throughput figures for the RabbitMQ/Redis pipeline under representative loads. These additions will allow evaluation of overfitting risk, domain shift, and false-positive rates relevant to real-time alerting. revision: yes

Circularity Check

0 steps flagged

No significant circularity in derivation chain

full rationale

The manuscript is a high-level system proposal describing a microservices architecture (RabbitMQ ingestion, Redis caching, modular services) and states that BERT-based models achieve 87% average accuracy on hate-speech detection. No equations, derivations, fitted parameters, or mathematical predictions appear in the provided text. There are no self-definitional steps, no claims that a prediction reduces to its own inputs by construction, and no load-bearing self-citations. The accuracy figure is presented as an experimental result without supporting dataset, training, or validation details, but this is an evidence gap rather than circularity. The architecture and claims remain independent of any self-referential reduction.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

The abstract introduces no free parameters, axioms, or invented entities; the platform relies on standard off-the-shelf components whose properties are taken from prior literature.

pith-pipeline@v0.9.0 · 5496 in / 1216 out tokens · 43907 ms · 2026-05-14T20:49:29.313053+00:00 · methodology

Review history (2 revisions) →

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

130 extracted references · 63 canonical work pages · 1 internal anchor

  1. [1]

    Leithardt, V. R. Q., et al. (2010). Use of Data Replication in WSNs Directed to Special Needs. DOI: 10.1109/ICCW.2010.5503914

    work page doi:10.1109/iccw.2010.5503914 2010

  2. [2]

    Noetzold, D., et al. (2026). Self-Adaptive Systems in Smart Environments: Strategies, Metrics, and Challenges. DOI: 10.1109/MC.2025.3637543

    work page doi:10.1109/mc.2025.3637543 2026

  3. [3]

    Dafhalla, A. K. Y., et al. (2025). Innovative Horizons in IoT and Cloud Continuum: Enhancing Data Processing and Advanced Analytical Solutions. Journal of Computer Networks and Communications, Article ID 3635533, 20 pages. DOI: 10.1155/jcnc/3635533

    work page doi:10.1155/jcnc/3635533 2025

  4. [4]

    R., et al

    Reshma, C. R., et al. (2025). Dynamic Sensitivity Differential Privacy: A Versatile Framework for Enhancing Privacy Across Diverse Data Scenarios. Journal of Computer Networks and Communications, Article ID 2972993, 18 pages. DOI: 10.1155/jcnc/2972993

    work page doi:10.1155/jcnc/2972993 2025

  5. [5]

    Leithardt, V. R. Q., et al. (2014). Privacy management solution in ubiquitous environments using percontrol. In: Journal of Ubiquitous Systems & Pervasive Networks, vol. 5, no. 2, pp. 21--28

    2014

  6. [6]

    Emon, M. J. H., et al. (2025). Integration of IoT, Machine Learning, and Sensors for Intelligent Environmental Monitoring and Agricultural Development. Journal of Computer Networks and Communications, Article ID 6611890, 20 pages. DOI: 10.1155/jcnc/6611890

    work page doi:10.1155/jcnc/6611890 2025

  7. [7]

    Pinto, J. (2023). Spearmint: A Decentralized Solution for Recording Evidence of Cyberbullying. CISTI 2023, pp. 110--122. DOI: 10.23919/CISTI58278.2023.10211594

    work page doi:10.23919/cisti58278.2023.10211594 2023

  8. [8]

    Noetzold, D., et al. (2026). Reinforcement Learning-Based Adaptive Quantum-Safe Cryptography for DN25-Compliant Smart Environments. DOI: 10.1109/ACCESS.2026.3685890

    work page doi:10.1109/access.2026.3685890 2026

  9. [9]

    Leithardt, V. R. Q., et al. (2014). Privacy Management Solution in Ubiquitous Environments Using Percontrol. Journal of Ubiquitous Systems & Pervasive Networks, vol. 5, no. 2, pp. 21--28. Available at: https://www.jusat.org/index.php/jusat/article/view/133 . (Accessed in: 17 Jan 2026)

    2014

  10. [10]

    Zhang, Y., et al. (2021). Hate crimes against Asian Americans. American Journal of Criminal Justice, pp. 1--21. DOI: 10.1007/s12103-020-09602-9

    work page doi:10.1007/s12103-020-09602-9 2021

  11. [11]

    Said-Hung, E., et al. (2024). The Promotion of Hate Speech: From a Media and Journalism Perspective. Journalism Practice, vol. 18, no. 2, pp. 217--223. DOI: 10.1080/17512786.2023.2288918

    work page doi:10.1080/17512786.2023.2288918 2024

  12. [12]

    Samuel, R. A. M., et al. (2024). Investigation of Data Interception and its Vulnerabilities in HTTP and HTTPS networks through Packet Sniffers. ICNWC 2024, pp. 1--4. DOI: 10.1109/ICNWC60771.2024.10537403

    work page doi:10.1109/icnwc60771.2024.10537403 2024

  13. [13]

    Smith, J., & Johnson, A. (2022). A Comparative Study: Undertow vs. Tomcat for Web Application Performance. Journal of Web Engineering, vol. 10, no. 2, pp. 45--60. DOI: 10.1234/jwe.2022.10.2.45

    work page doi:10.1234/jwe.2022.10.2.45 2022

  14. [14]

    Granata, D., et al. (2023). Attacking a Packet Analyzer: Caronte case study. Proceedings of CSR 2023, pp. 269--274. DOI: 10.1109/CSR57506.2023.10224918

    work page doi:10.1109/csr57506.2023.10224918 2023

  15. [15]

    A., et al

    Glasenapp, L. A., et al. (2023). OCR applied for identification of vehicles with irregular documentation using IoT. Electronics, vol. 12, no. 5, p. 1083. DOI: 10.3390/electronics12051083

    work page doi:10.3390/electronics12051083 2023

  16. [17]

    Pinto, J. L. D., et al. (2023). Spearmint: A Decentralized Solution for Recording Evidence of Cyberbullying. In: New Trends in Disruptive Technologies, Tech Ethics and Artificial Intelligence. Springer, pp. 110--122. Available at: https://link.springer.com/chapter/10.1007/978-3-031-38345-8\_10 . (Accessed in: 17 Jan 2026)

    work page doi:10.1007/978-3-031-38345-8 2023

  17. [18]

    Assessing the Impact of Contextual Information in Hate Speech Detection

    Perez (2023). Assessing the Impact of Contextual Information in Hate Speech Detection. IEEE Access, vol. 11, pp. 30575--30590. DOI: 10.1109/ACCESS.2023.3258973

    work page doi:10.1109/access.2023.3258973 2023

  18. [19]

    Ribeiro, M. H. D. M., et al. (2020). Electricity Price Forecasting Based on Self-Adaptive Decomposition and Heterogeneous Ensemble Learning. Energies, vol. 13, no. 19. DOI: 10.3390/en13195190

    work page doi:10.3390/en13195190 2020

  19. [20]

    R., et al

    Moreno, S. R., et al. (2021). Hybrid Multi-Stage Decomposition with Parametric Model Applied to Wind Speed Forecasting in Brazilian Northeast. Renewable Energy, vol. 164, pp. 1508--1526. DOI: 10.1016/j.renene.2020.10.126

    work page doi:10.1016/j.renene.2020.10.126 2021

  20. [21]

    Jury Orders Tesla to Pay \ 137 Million to a Former Worker Over Racist Treatment

    Nytime (2021). Jury Orders Tesla to Pay \ 137 Million to a Former Worker Over Racist Treatment. Available at: https://www.nytimes.com/2021/10/04/business/tesla-racism-lawsuit.html . (Accessed in: 17 Jan 2026)

    2021

  21. [22]

    Cost of a Data Breach 2022

    IBM (2022). Cost of a Data Breach 2022. Available at: https://www.ibm.com/downloads/cas/NNZMWXZL . (Accessed in: 17 Jan 2026)

    2022

  22. [23]

    F., et al

    Stefenon, S. F., et al. (2023a). Optimized Hybrid YOLOu-Quasi-ProtoPNet for Insulators Classification. IET Generation, Transmission & Distribution, vol. 17, no. 15, pp. 3501--3511. DOI: 10.1049/gtd2.12886

    work page doi:10.1049/gtd2.12886

  23. [24]

    J., et al

    Souza, B. J., et al. (2023). Hybrid-YOLO for Classification of Insulators Defects in Transmission Lines Based on UAV. International Journal of Electrical Power & Energy Systems, vol. 148, p. 108982. DOI: 10.1016/j.ijepes.2023.108982

    work page doi:10.1016/j.ijepes.2023.108982 2023

  24. [25]

    J., et al

    Wilbert, H. J., et al. (2023). Recency, Frequency, Monetary Value, Clustering, and Internal and External Indices for Customer Segmentation from Retail Data. Algorithms, vol. 16, no. 9, p. 396. DOI: 10.3390/a16090396

    work page doi:10.3390/a16090396 2023

  25. [26]

    Klaar, A. C. R., et al. (2023a). Structure Optimization of Ensemble Learning Methods and Seasonal Decomposition Approaches to Energy Price Forecasting in Latin America: A Case Study About Mexico. Energies, vol. 16, no. 7, p. 3184. DOI: 10.3390/en16073184

    work page doi:10.3390/en16073184

  26. [27]

    Klaar, A. C. R., et al. (2023b). Optimized EWT-Seq2Seq-LSTM with Attention Mechanism to Insulators Fault Prediction. Sensors, vol. 23, no. 6, p. 3202. DOI: 10.3390/s23063202

    work page doi:10.3390/s23063202

  27. [28]

    2020 Hate Crime Data Now Available! Available at: https://hatecrime.osce.org/infocus/2020-hate-crime-data-now-available

    OSCE (2021). 2020 Hate Crime Data Now Available! Available at: https://hatecrime.osce.org/infocus/2020-hate-crime-data-now-available . (Accessed in: 17 Jan 2026)

    2021

  28. [29]

    DeskTime

    DeskTime (2024). DeskTime. Available at: https://desktime.com . (Accessed in: 17 Jan 2026)

    2024

  29. [30]

    Linguistic Datasets for Portuguese

    fititnt (2019). Linguistic Datasets for Portuguese. Available at: https://github.com/EticaAI/linguistic-datasets-portuguese . (Accessed in: 17 Jan 2026)

    2019

  30. [31]

    Kansaon, D. (2018). BraSNAM2018-Dataset-Analise-de-sentimentos-em-tweets-em-portugues-brasileiro. Available at: https://github.com/danielkansaon/BraSNAM2018-Dataset-Analise-de-sentimentos-em-tweets-em-portugues-brasileiro . (Accessed in: 17 Jan 2026)

    2018

  31. [32]

    Staffcop

    Atom Security LLC (2024). Staffcop. Available at: https://www.staffcop.com . (Accessed in: 17 Jan 2026)

    2024

  32. [33]

    Programa Para Monitorar e Controlar Computadores de Funcionários

    Kickidler (2024). Programa Para Monitorar e Controlar Computadores de Funcionários. Available at: https://www.kickidler.com/br . (Accessed in: 17 Jan 2026)

    2024

  33. [34]

    Workforce Analytics for Productivity Management

    ActivTrak (2023). Workforce Analytics for Productivity Management. Available at: https://www.activtrak.com . (Accessed in: 17 Jan 2026)

    2023

  34. [35]

    fSense: Sistema de Monitoramento Prático e Preciso para Estações de Trabalho

    FSense (2023). fSense: Sistema de Monitoramento Prático e Preciso para Estações de Trabalho. Available at: https://fsense.com . (Accessed in: 17 Jan 2026)

    2023

  35. [36]

    Paschalides, D. (2020). Mandola: A Big-Data Processing and Visualization Platform for Monitoring and Detecting Online Hate Speech. ACM Transactions on Internet Technology (TOIT), vol. 20, no. 2, pp. 1--21. DOI: 10.1145/3371276

    work page doi:10.1145/3371276 2020

  36. [37]

    Gangwani, P., et al. (2026). Evaluating Convolutional Autoencoders for Anomaly Detection on Space-Filling Curve-Transformed Control Flow Data. IEEE Access, vol. 14, pp. 4292--4304. DOI: 10.1109/ACCESS.2026.3651178

    work page doi:10.1109/access.2026.3651178 2026

  37. [38]

    Ababneh, J., et al. (2026). Enhancing DevOps Continuous Monitoring Phase: Hybrid Intrusion Detection and Ensemble Learning System (HIDELS). IEEE Access, vol. 14, pp. 4733--4755. DOI: 10.1109/ACCESS.2026.3650793

    work page doi:10.1109/access.2026.3650793 2026

  38. [39]

    Noetzold, D., et al. (2023). DN25: An Adaptive Quantum Cryptography Protocol for Secure and Efficient Communication. Revista de I+D Tecnológico, vol. 19, no. 1. Available at: https://revistas.unilasalle.edu.br/index.php/idt/article/view/8429 . (Accessed in: 17 Jan 2026)

    2023

  39. [40]

    Modha, S. (2020). Detecting and Visualizing Hate Speech in Social Media: A Cyber Watchdog for Surveillance. Expert Systems with Applications, vol. 161, p. 113725. DOI: 10.1016/j.eswa.2020.113725

    work page doi:10.1016/j.eswa.2020.113725 2020

  40. [41]

    O., et al

    Seman, L. O., et al. (2023). Ensemble Learning Methods Using the Hodrick--Prescott Filter for Fault Forecasting in Insulators of the Electrical Power Grids. International Journal of Electrical Power & Energy Systems, vol. 152, p. 109269. DOI: 10.1016/j.ijepes.2023.109269

    work page doi:10.1016/j.ijepes.2023.109269 2023

  41. [42]

    Yamasaki, M., et al. (2024). Optimized Hybrid Ensemble Learning Approaches Applied to Very Short-Term Load Forecasting. International Journal of Electrical Power & Energy Systems, vol. 155, p. 109579. DOI: 10.1016/j.ijepes.2023.109579

    work page doi:10.1016/j.ijepes.2023.109579 2024

  42. [43]

    F., et al

    Stefenon, S. F., et al. (2023b). Aggregating Prophet and Seasonal Trend Decomposition for Time Series Forecasting of Italian Electricity Spot Prices. Energies, vol. 16, no. 3, p. 1371. DOI: 10.3390/en16031371

    work page doi:10.3390/en16031371

  43. [44]

    Begnum, M., & Vinterhagen, A. (2020). Evaluating the Redis In-Memory Cache for Distributed Storage Systems. Journal of Computer and Communications, vol. 8, no. 12, pp. 1--15. DOI: 10.4236/jcc.2020.812001

    work page doi:10.4236/jcc.2020.812001 2020

  44. [45]

    Gao, P., et al. (2018). Performance Evaluation of Redis Cache in Web Service. 2018 IEEE International Conference on Communications Workshops (ICC Workshops), pp. 1--6. DOI: 10.1109/ICCW.2018.8403584

    work page doi:10.1109/iccw.2018.8403584 2018

  45. [46]

    E., et al

    Bagaskara, A. E., et al. (2020). Performance Analysis of Message Broker for Communication in Fog Computing. ICITEE 2020, pp. 98--103. DOI: 10.1109/ICITEE49829.2020.9271733

    work page doi:10.1109/icitee49829.2020.9271733 2020

  46. [47]

    Abdelwahab, S., & Hamdaoui, B. (2016). Fogmq: A Message Broker System for Enabling Distributed, Internet-Scale IoT Applications Over Heterogeneous Cloud Platforms. arXiv preprint arXiv:1610.00620. DOI: 10.48550/arXiv.1610.00620

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1610.00620 2016

  47. [48]

    L., et al

    de Oliveira, D. L., et al. (2019). Performance Evaluation of MQTT Brokers in the Internet of Things for Smart Cities. SpliTech 2019, pp. 1--6. DOI: 10.23919/SpliTech.2019.8783166

    work page doi:10.23919/splitech.2019.8783166 2019

  48. [49]

    R., et al

    Moreno, S. R., et al. (2024). Enhancing Wind Speed Forecasting Through Synergy of Machine Learning, Singular Spectral Analysis, and Variational Mode Decomposition. Energy, vol. 292, p. 130493. DOI: 10.1016/j.energy.2024.130493

    work page doi:10.1016/j.energy.2024.130493 2024

  49. [50]

    Fernandes, F., et al. (2022). Long Short-Term Memory Stacking Model to Predict the Number of Cases and Deaths Caused by COVID-19. Journal of Intelligent & Fuzzy Systems, vol. 42, no. 6, pp. 6221--6234. DOI: 10.3233/JIFS-212788

    work page doi:10.3233/jifs-212788 2022

  50. [51]

    F., et al

    Stefenon, S. F., et al. (2022). Analysis of the Ultrasonic Signal in Polymeric Contaminated Insulators Through Ensemble Learning Methods. IEEE Access, vol. 10, pp. 33980--33991. DOI: 10.1109/ACCESS.2022.3161506

    work page doi:10.1109/access.2022.3161506 2022

  51. [52]

    F., et al

    Stefenon, S. F., et al. (2021). Photovoltaic Power Forecasting Using Wavelet Neuro-Fuzzy for Active Solar Trackers. Journal of Intelligent & Fuzzy Systems, vol. 40, no. 1, pp. 1083--1096. DOI: 10.3233/JIFS-201279

    work page doi:10.3233/jifs-201279 2021

  52. [53]

    F., et al

    Sopelsa Neto, N. F., et al. (2022). Fault Prediction Based on Leakage Current in Contaminated Insulators Using Enhanced Time Series Forecasting Models. Sensors, vol. 22, no. 16, p. 6121. DOI: 10.3390/s22166121

    work page doi:10.3390/s22166121 2022

  53. [54]

    F., et al

    Stefenon, S. F., et al. (2019). Fault Diagnosis of Insulators From Ultrasound Detection Using Neural Networks. Journal of Intelligent & Fuzzy Systems, vol. 37, no. 5, pp. 6655--6664. DOI: 10.3233/JIFS-190013

    work page doi:10.3233/jifs-190013 2019

  54. [55]

    Fersini, E., et al. (2018). Overview of the Task on Automatic Misogyny Identification at IberEval 2018. CEUR-WS, vol. 2150. Available at: https://ceur-ws.org/Vol-2150/overview-AMI.pdf . (Accessed in: 17 Jan 2026)

    2018

  55. [56]

    A., et al

    Álvarez-Carmona, M. A., et al. (2018). Overview of MEX-A3T at IberEval 2018: Authorship and Aggressiveness Analysis in Mexican Spanish Tweets. CEUR-WS, vol. 2150. Available at: https://ceur-ws.org/Vol-2150/overview-mex-a3t.pdf . (Accessed in: 17 Jan 2026)

    2018

  56. [57]

    Dadvar, M., et al. (2012). Improved Cyberbullying Detection Using Gender Information. Available at: https://www.researchgate.net/publication/230701861\_Improved\_Cyberbullying \_Detection\_Using\_Gender\_Information . (Accessed in: 17 Jan 2026)

    work page arXiv 2012

  57. [58]

    Waseem, Z. (2016). Are You a Racist or Am I Seeing Things? Annotator Influence on Hate Speech Detection on Twitter. Available at: https://aclanthology.org/W16-5618 . (Accessed in: 17 Jan 2026)

    2016

  58. [59]

    Zulqarnain, M., et al. (2020). A Comparative Review on Deep Learning Models for Text Classification. Indonesian Journal of Electrical Engineering and Computer Science, vol. 19, no. 1, pp. 325--335. DOI: 10.11591/ijeecs.v19.i1.pp325-335

    work page doi:10.11591/ijeecs.v19.i1.pp325-335 2020

  59. [60]

    Hastie, T., et al. (2009). The Elements of Statistical Learning. Springer. DOI: 10.1198/tech.2003.s770

    work page doi:10.1198/tech.2003.s770 2009

  60. [61]

    Cortes, C., & Vapnik, V. (1995). Support-Vector Networks. Machine Learning, vol. 20, pp. 273--297. Available at: https://ise.ncsu.edu/wp-content/uploads/sites/9/2022/08/Cortes-Vapnik1995\_Article\_Support-vectorNetworks.pdf . (Accessed in: 17 Jan 2026)

    1995

  61. [62]

    Chen, X., et al. (2001). An Admission Control Scheme for Predictable Server Response Time for Web Accesses. Proc. SIGMETRICS, pp. 545--554. DOI: 10.1145/371920.372156

    work page doi:10.1145/371920.372156 2001

  62. [63]

    Kalischko, T., & Riedl, R. (2021). Electronic Performance Monitoring in the Digital Workplace: Conceptualization, Review of Effects and Moderators, and Future Research Opportunities. Frontiers in Psychology, vol. 12, p. 633031. DOI: 10.3389/fpsyg.2021.633031

    work page doi:10.3389/fpsyg.2021.633031 2021

  63. [64]

    Sherif, K., et al. (2021). Empowering Employees: the Other Side of Electronic Performance Monitoring. Journal of Information, Communication and Ethics in Society, vol. 19, no. 2, pp. 207--221. DOI: 10.1108/JICES-04-2020-0038

    work page doi:10.1108/jices-04-2020-0038 2021

  64. [65]

    Wieser, L., et al. (2023). When Are Employers Interested in Electronic Performance Monitoring? Results from a Factorial Survey Experiment. IZA Discussion Paper No. 16694. Available at: https://www.iza.org/publications/dp/16694 . (Accessed in: 17 Jan 2026)

    2023

  65. [66]

    Channi, H. K. (2024). Role of Data Acquisition, Sensors, and Actuators in Mechatronics Industry. Computational Intelligent Techniques in Mechatronics, pp. 83--107. DOI: 10.1002/9781394175437.ch3

    work page doi:10.1002/9781394175437.ch3 2024

  66. [67]

    Kilic, I., & Yaman, O. (2024). Classification of Spyware from Network Packets with Decision Trees Using Recursive Feature Elimination (RFE). pp. 1--4. DOI: 10.1109/SIU61531.2024.10600885

    work page doi:10.1109/siu61531.2024.10600885 2024

  67. [68]

    Rossetto, A. G. M., et al. (2024). Enhancing Monitoring Performance: A Microservices Approach to Monitoring with Spyware Techniques and Prediction Models. Sensors, vol. 24, no. 13. DOI: 10.3390/s24134212

    work page doi:10.3390/s24134212 2024

  68. [69]

    Krieter, P., et al. (2024). Following Topics Across All Apps and Media Formats: Mobile Keyword Tracking as a Privacy-Friendly Data Source in Mobile Media Research. pp. 126--131. DOI: 10.1145/3631700.3664879

    work page doi:10.1145/3631700.3664879 2024

  69. [70]

    S., et al

    Malik, J. S., et al. (2024). Deep learning for hate speech detection: a comparative study. International Journal of Data Science and Analytics, vol. 9, no. 4, pp. 123--139. DOI: 10.1007/s41060-024-00650-6

    work page doi:10.1007/s41060-024-00650-6 2024

  70. [71]

    Abu Bakar, R., & Kijsirikul, B. (2023). Enhancing Network Visibility and Security with Advanced Port Scanning Techniques. Sensors, vol. 23, no. 17. DOI: 10.3390/s23177541

    work page doi:10.3390/s23177541 2023

  71. [72]

    Noetzold, D. (2024). spyware-API. Available at: https://github.com/DarlanNoetzold/spyware-API . (Accessed in: 17 Jan 2026)

    2024

  72. [73]

    D., et al

    Samudera, B. D., et al. (2024). Sentiment Analysis of User Reviews on BSI Mobile and Action Mobile Applications on the Google Play Store Using Multinomial Naive Bayes Algorithm. International Journal of Engineering, Science and Information Technology, vol. 4, no. 4, pp. 101--112. DOI: 10.52088/ijesty.v4i4.581

    work page doi:10.52088/ijesty.v4i4.581 2024

  73. [74]

    Javaheri, D., et al. (2024). Cybersecurity threats in FinTech: A systematic review. Expert Systems with Applications, vol. 241, p. 122697. DOI: 10.1016/j.eswa.2023.122697

    work page doi:10.1016/j.eswa.2023.122697 2024

  74. [75]

    F., et al

    Heckler, W. F., et al. (2023). Thoth: An intelligent model for assisting individuals with suicidal ideation. Expert Systems with Applications, vol. 233, p. 120918. DOI: 10.1016/j.eswa.2023.120918

    work page doi:10.1016/j.eswa.2023.120918 2023

  75. [76]

    Wulandari, S. (2023). Spyware-Based Keylogger Design for Monitoring Smartphone User Activities Using the Smart Keylogger Application. Progressive Information, Security, Computer, and Embedded System, vol. 1, no. 2, pp. 149--157. DOI: 10.61255/pisces.v1i2.98

    work page doi:10.61255/pisces.v1i2.98 2023

  76. [77]

    Ramos, G., et al. (2024). A comprehensive review on automatic hate speech detection in the age of the transformer. Social Network Analysis and Mining, vol. 14, no. 204. DOI: 10.1007/s13278-024-01361-3

    work page doi:10.1007/s13278-024-01361-3 2024

  77. [78]

    Zheng, Z., et al. (2010). Distributed QoS Evaluation for Real-World Web Services. pp. 83--90. DOI: 10.1109/ICWS.2010.10

    work page doi:10.1109/icws.2010.10 2010

  78. [79]

    Aydemir, O. (2020). A New Performance Evaluation Metric for Classifiers: Polygon Area Metric. Journal of Classification. DOI: 10.1007/s00357-020-09362-5

    work page doi:10.1007/s00357-020-09362-5 2020

  79. [80]

    Chen, L., et al. (2024). Uncertainty-Aware Cross-Modal Alignment for Hate Speech Detection. Available at: https://aclanthology.org/2024.lrec-main.1475/ . (Accessed in: 17 Jan 2026)

    2024

  80. [81]

    Smith, J., et al. (2024). PREDICT: Multi-Agent-based Debate Simulation for Generalized Hate Speech Detection. Available at: https://aclanthology.org/2024.emnlp-main.1166/ . (Accessed in: 17 Jan 2026)

    2024

Showing first 80 references.