Enhanced security and privacy in IoMT: a hierarchical federated learning approach using Dew-Cloud with HLSTM for hostile attack mitigation
Bibliographic record
Abstract
Due to the coronavirus pandemic, doctors have had to treat patients remotely while medical facilities are overwhelmed. Also, as a result of COVID-19, people are far more concerned about their health, which has increased demand for internet-connected medical devices. Because of its incredible growth in value, the internet of medical things (IoMT) has caught the attention of cybercriminals. Many people's private health data and other very sensitive documents are safe on the dark web. Regardless, the trespassers were able to take advantage of the patient's health information because it was not adequately protected. The system administrator cannot tighten security since resource-constrained network devices do not have enough space or processing power. The primary objective is to study the expanding hostile attacks before they jeopardize the health system's security, while there are several supervised and unsupervised machine learning techniques that can detect outliers. This study's methodology utilizes Dew-Cloud to provide hierarchical federated learning (HFL). More availability of critical IoMT application(s) and enhanced data privacy are two advantages of the proposed Dew-Cloud idea. The hierarchical LSTM (long short-term memory) concept is used by distributed Dew servers that employ cloud computing for their backend implementation. Training the proposed model with the data pre-processing feature results in a low loss of 0.034 and a high accuracy of 99.31%. The suggested HFL-HLSTM model surpasses other methods in a number of performance parameters, such as f-score, recall, accuracy, and precision.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.004 | 0.012 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".