Knowledge, Volume 4, Issue 3 (September 2024) – 6 articles

Individuals from a variety of backgrounds are affected by the opioid crisis. To provide optimal care for individuals at risk of opioid overdose and prevent subsequent overdoses, a more _targeted response that goes beyond the traditional taxonomical diagnosis approach to care management needs to be adopted. In previous works, Graph Machine Learning and Natural Language Processing methods were used to model the products for planning and evaluating the treatment of patients with complex issues. This study proposes a methodology of partitioning patients in the opioid overdose cohort into various communities based on their patterns of service utilization (PSUs) across the continuum of care using graph community detection and applying survival analysis to predict time-to-second overdose for each of the communities. The results demonstrated that the overdose cohort is not homogeneous with respect to the determinants of risk. Moreover, the risk for subsequent overdose was quantified: there is a 51% higher chance of experiencing a second overdose for a high-risk community compared to a low-risk community. The proposed method can inform a more efficient treatment heterogeneity approach for a cohort made of diverse individuals, such as the opioid overdose cohort. It can also guide _targeted support for patients at risk of subsequent overdoses. Full article

Variations in the genetic code for proteins are considered to confer traits and underlying disease. Identifying the functional consequences of these genetic variants is a challenging endeavor. There are online databases that contain variant information. Many publications also have described variants in detail. Furthermore, there are tools that allow for the prediction of the pathogenicity of variants. However, navigating these disparate sources is time-consuming and sometimes complex. Finally, text mining and large language models offer promising approaches to understanding the textual form of this knowledge. This review discusses these challenges and the online resources and tools available to facilitate this process. Furthermore, a computational framework is suggested to accelerate and facilitate the process of identifying the phenotype caused by a particular genetic variant. This framework demonstrates a way to gather and understand the knowledge about variants more efficiently and effectively. Full article

This paper introduces a parameter-efficient transformer-based model designed for scientific literature classification. By optimizing the transformer architecture, the proposed model significantly reduces memory usage, training time, inference time, and the carbon footprint associated with large language models. The proposed approach is evaluated against various deep learning models and demonstrates superior performance in classifying scientific literature. Comprehensive experiments conducted on datasets from Web of Science, ArXiv, Nature, Springer, and Wiley reveal that the proposed model’s multi-headed attention mechanism and enhanced embeddings contribute to its high accuracy and efficiency, making it a robust solution for text classification tasks. Full article

Undergraduate courses in electrical machines often include an introduction to the air gap magnetic field as a basic element in the energy conversion process. The students must learn the main properties of the field produced by basic winding configurations and how they relate to the winding current and frequency. This paper describes a new test equipment design aimed at helping students achieve these learning goals. The test equipment is designed based on four main elements: a modified slip ring induction machine, a winding current driver board, the DAQ boards, and a PC-based virtual instrument. The virtual instrument provides the winding current drivers with suitable current references depending on the user selected machine operational status (single- or three-phase/winding with DC or AC current) and measures and displays the air gap magnetic field for that operational status. Students’ laboratory work is organized into a series of experiments that guide their achievement of these air gap field-related abilities. Student learning, assessed based on pre- and post-lab exams and end-of-semester exams, has increased significantly. The students’ opinions of the relevance, usefulness, and motivational effects of the laboratory were also positive. Full article

In response to the recent formalization of Filipino Sign Language (FSL) and the lack of comprehensive studies, this paper introduces a real-time FSL gesture recognition system. Unlike existing systems, which are often limited to static signs and asynchronous recognition, it offers dynamic gesture capturing and recognition of 10 common expressions and five transactional inquiries. To this end, the system sequentially employs cropping, contrast adjustment, grayscale conversion, resizing, and normalization of input image streams. These steps serve to extract the region of interest, reduce the computational load, ensure uniform input size, and maintain consistent pixel value distribution. Subsequently, a Convolutional Neural Network and Long-Short Term Memory (CNN-LSTM) model was employed to recognize nuances of real-time FSL gestures. The results demonstrate the superiority of the proposed technique over existing FSL recognition systems, achieving an impressive average accuracy, recall, and precision rate of 98%, marking an 11.3% improvement in accuracy. Furthermore, this article also explores lightweight conversion methods, including post-quantization and quantization-aware training, to facilitate the deployment of the model on resource-constrained platforms. The lightweight models show a significant reduction in model size and memory utilization with respect to the base model when executed in a Raspberry Pi minicomputer. Lastly, the lightweight model trained with the quantization-aware technique (99%) outperforms the post-quantization approach (97%), showing a notable 2% improvement in accuracy. Full article

We develop a new theory of knowledge with mathematics and a broad-based series of case studies to seek a better understanding of what constitutes knowledge in the field and its value for autonomous human–machine teams facing uncertainty in the open. Like humans, as teammates, artificial intelligence (AI) machines must be able to determine what constitutes the usable knowledge that contributes to a team’s success when facing uncertainty in the field (e.g., testing “knowledge” in the field with debate; identifying new knowledge; using knowledge to innovate), its failure (e.g., troubleshooting; identifying weaknesses; discovering vulnerabilities; exploitation using deception), and feeding the results back to users and society. It matters not whether a debate is public, private, or unexpressed by an individual human or machine agent acting alone; regardless, in this exploration, we speculate that only a transparent process advances the science of autonomous human–machine teams, assists in interpretable machine learning, and allows a free people and their machines to co-evolve. The complexity of the team is taken into consideration in our search for knowledge, which can also be used as an information metric. We conclude that the structure of “knowledge”, once found, is resistant to alternatives (i.e., it is ordered); that its functional utility is generalizable; and that its useful applications are multifaceted (akin to maximum entropy production). Our novel finding is the existence of Shannon holes that are gaps in knowledge, a surprising “discovery” to only find Shannon there first. Full article