Ontology-based semantic data interestingness using BERT models

The COVID-19 pandemic has generated massive data in the healthcare sector in recent years, encouraging researchers and scientists to uncover the underlying facts. Mining interesting patterns in the large COVID-19 corpora is very important and useful for the decision makers. This paper presents a novel approach for uncovering interesting insights in large datasets using ontologies and BERT models. The research proposes a framework for extracting semantically rich facts from data by incorporating domain knowledge into the data mining process through the use of ontologies. An improved Apriori algorithm is employed for mining semantic association rules, while the interestingness of the rules is evaluated using BERT models for semantic richness. The results of the proposed framework are compared with state-of-the-art methods and evaluated using a combination of domain expert evaluation and statistical significance testing. The study offers a promising solution for finding meaningful relationships and facts in large datasets, particularly in the healthcare sector. © 2023 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.

Preprocessing and Artificial Intelligence for Increasing Explainability in Mental Health

This paper shows the added value of using the existing specific domain knowledge to generate new derivated variables to complement a target dataset and the benefits of including these new variables into further data analysis methods. The main contribution of the paper is to propose a methodology to generate these new variables as a part of preprocessing, under a double approach: creating 2nd generation knowledge-driven variables, catching the experts criteria used for reasoning on the field or 3rd generation data-driven indicators, these created by clustering original variables. And Data Mining and Artificial Intelligence techniques like Clustering or Traffic light Panels help to obtain successful results. Some results of the project INSESS-COVID19 are presented, basic descriptive analysis gives simple results that even though they are useful to support basic policy-making, especially in health, a much richer global perspective is acquired after including derivated variables. When 2nd generation variables are available and can be introduced in the method for creating 3rd generation data, added value is obtained from both basic analysis and building new data-driven indicators. © 2023 World Scientific Publishing Company.

Mining Literature-Based Knowledge Graph for Predicting Combination Therapeutics: A COVID-19 Use Case

This paper presents a computational approach designed to construct and query a literature-based knowledge graph for predicting novel drug therapeutics. The main objective is to offer a platform that discovers drug combinations from FDA-approved drugs and accelerates their investigations by domain scientists. Specifically, the paper introduced the following algorithms: (1) an algorithm for constructing the knowledge graph from drug, gene, and disease mentions in the biomedical literature;(2) an algorithm for vetting the knowledge graph from drug combinations that may pose a risk of drug interaction;(3) and two querying algorithms for searching the knowledge graph by a single drug or a combination of drugs. The resulting knowledge graph had 844 drugs, 306 gene/protein features, and 19 disease mentions. The original number of drug combinations generated was 2,001. We queried the knowledge graph to eliminate noise generated from chemicals that are not drugs. This step resulted in 614 drug combinations. When vetting the knowledge graph to eliminate the potentially risky drug combinations, it resulted in predicting 200 combinations. Our domain expert manually eliminated extra 54 combinations which left only 146 combination candidates. Our three-layered knowledge graph, empowered by our algorithms, offered a tool that predicted drug combination therapeutics for scientists who can further investigate from the viewpoint of drug targets and side effects. © 2022 IEEE.

Using a Layered Ensemble of Physics-Guided Graph Attention Networks to Predict COVID-19 Trends

The COVID-19 pandemic has spread rapidly and significantly impacted most countries in the world. Providing an accurate forecast of COVID-19 at multiple scales would help inform public health decisions, but recent forecasting models are typically used at the state or country level. Furthermore, traditional mathematical models are limited by simplifying assumptions, while machine learning algorithms struggle to generalize to unseen trends. This motivates the need for hybrid machine learning models that integrate domain knowledge for accurate long-term prediction. We propose a three-layer, geographically informed ensemble, an extensive peer-learning framework, for predicting COVID-19 trends at the country, continent, and global levels. As the base layer, we develop a country-level predictor using a hybrid Graph Attention Network that incorporates a modified SIR model, adaptive loss function, and edge weights informed by mobility data. We aggregated 163 country GATs to train the continent and world MLP layers of the ensemble. Our results indicate that incorporating quantitatively accurate equations and real-world data to model inter-community interactions improves the performance of spatio-temporal machine learning algorithms. Additionally, we demonstrate that integrating geographic information (continent composition) improves the performance of the world predictor in our layered architecture. (PsycInfo Database Record (c) 2022 APA, all rights reserved)

Bridging the Gap Between Domain Models and Computational Models: A Case Study of COVID-19

Novel coronavirus pneumonia (COVID-19) has broken out and spread rapidly in many countries and regions around the world. Since the outbreak, many researchers have proposed propagation models of COVID-19, among which the mainstream computational epidemiology model requires the establishment of a corresponding artificial society model for computational experiments. However, such models tightly coupled domain knowledge about epidemics with computational models and have low reusability. On this basis, we take COVID-19 as our research object and propose a hierarchical modeling framework for epidemic transmission, which describes how to decouple and dock domain models and computational models. This framework consists of three levels: individual capability model and virus model at the individual level, organizational structure and interaction mechanisms between individuals at the organizational level, and intervention model and environmental model design at the social level. The experimental results show that this is an effective hierarchical framework modeling approach for studying transmission mechanisms. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Automatic Exploration of Domain Knowledge in Healthcare

Throughout the years, healthcare has been one of the privileged areas to apply the information discovery process, empowering and supporting medical staff on their daily activities. One of the main reasons for its success is the availability of medical expertise, which can be incorporated in training models to reach higher levels of performance. While this has been done painfully and manually, during the preparation step, it has become hindered with the advent of AutoML. In this paper, we present the automation of data preparation and feature engineering, while exploring domain knowledge represented through extended entity-relationship (EER) diagrams. A COVID-19 case study shows that our automation outperforms existing AutoML tools, such as auto-sklearn [4], both in quality of the models and processing times. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Epidemic Forecasting with a Data-Centric Lens

The recent COVID-19 pandemic has reinforced the importance of epidemic forecasting to equip decision makers in multiple domains, ranging from public health to economics. However, forecasting the epidemic progression remains a non-trivial task as the spread of diseases is subject to multiple confounding factors spanning human behavior, pathogen dynamics and environmental conditions, etc. Research interest has been fueled by the increased availability of rich data sources capturing previously unseen facets of the epidemic spread and initiatives from government public health and funding agencies like forecasting challenges and funding calls. This has resulted in recent works covering many aspects of epidemic forecasting. Data-centered solutions have specifically shown potential by leveraging non-traditional data sources as well as recent innovations in AI and machine learning. This tutorial will explore various data-driven methodological and practical advancements. First, we will enumerate epidemiological datasets and novel data streams capturing various factors like symptomatic online surveys, retail and commerce, mobility and genomics data. Next, we discuss methods and modeling paradigms with a focus on the recent data-driven statistical and deep-learning based methods as well as novel class of hybrid models that combine domain knowledge of mechanistic models with the effectiveness and flexibility of statistical approaches. We also discuss experiences and challenges that arise in real-world deployment of these forecasting systems including decision-making informed by forecasts. Finally, we highlight some open problems found across the forecasting pipeline. © 2022 Owner/Author.

New Frontiers of Scientific Text Mining: Tasks, Data, and Tools

Exploring the vast amount of rapidly growing scientific text data is highly beneficial for real-world scientific discovery. However, scientific text mining is particularly challenging due to the lack of specialized domain knowledge in natural language context, complex sentence structures in scientific writing, and multi-modal representations of scientific knowledge. This tutorial presents a comprehensive overview of recent research and development on scientific text mining, focusing on the biomedical and chemistry domains. First, we introduce the motivation and unique challenges of scientific text mining. Then we discuss a set of methods that perform effective scientific information extraction, such as named entity recognition, relation extraction, and event extraction. We also introduce real-world applications such as textual evidence retrieval, scientific topic contrasting for drug discovery, and molecule representation learning for reaction prediction. Finally, we conclude our tutorial by demonstrating, on real-world datasets (COVID-19 and organic chemistry literature), how the information can be extracted and retrieved, and how they can assist further scientific discovery. We also discuss the emerging research problems and future directions for scientific text mining. © 2022 Owner/Author.

Rule Based Medicine Recommendation for Skin Diseases Using Ontology with Semantic Information

Generally, skin diseases are taken seriously only after it gets aggravated. Many patients do not feel comfortable to consult physician for their skin problems and try to cure with home remedies. Identifying good dermatologist is more challenging. Disease gets cured with ease, if treated properly otherwise it becomes complicated. To handle this, knowledge-based decision support system is developed which provide recommendation for treatment. System realizes patient’s symptoms with location, cause and recommends appropriate medicines. Ontology is used to elaborate data concepts and their relationships in skin disease, which permits sharing and reuse of domain knowledge. Semantics described using Web Ontology Language (OWL) with vocabularies, resources, logic’s, and inference rules are queried for relevant information through SPARQL. System proved to render most valuable service to skin patients with ease and accuracy of its recommendation is high. It renders best treatment in COVID situation where people’s movement is restricted to great extent. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

RES-Q: Toward Semantic Interoperability for Risk and Disaster Management in Smart Cities

The COVID-19 pandemic has imposed new challenges in preserving the goal of developing smart and sustainable cities worldwide while improving urban resilience. In the smart city domain, disaster or crisis management operations require contributions and collaboration from different types of entities with various functions, rules, and protocols, forming complex contexts in decision-making or event coordination. The management of the corresponding information usually coming from multiple heterogeneous sources and sometimes with attributes revealing semantic inconsistencies constitutes an emerging challenge. Furthermore, the demand for interoperability between the various services and IoT devices at local and national level is imperative. Yet, existing literature highlights that the conceptualization of a holistic reference schema that covers all the dimensions of the smart city disaster/crisis management domain and allows the exchange of information through different agents has not been fully addressed so far. We present the RES-Q (RESCUE) semantic model, which includes the needed domain knowledge streams for the smart city crisis management domain. This model aims for data consolidation and linkage in order to be further utilized for the implementation of a common knowledge repository and advanced analysis. In this context, semantic web technologies are proposed as a promising solution for providing semantic interoperability in crisis and/or disaster management in the smart city discourse. Finally, data consolidation and harmonization methodology is presented, which is used for the integration of different data sources, according to the RES-Q model. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Integrating Expert ODEs into Neural ODEs: Pharmacology and Disease Progression

Modeling a system's temporal behaviour in reaction to external stimuli is a fundamental problem in many areas. Pure Machine Learning (ML) approaches often fail in the small sample regime and cannot provide actionable insights beyond predictions. A promising modification has been to incorporate expert domain knowledge into ML models. The application we consider is predicting the patient health status and disease progression over time, where a wealth of domain knowledge is available from pharmacology. Pharmacological models describe the dynamics of carefully-chosen medically meaningful variables in terms of systems of Ordinary Differential Equations (ODEs). However, these models only describe a limited collection of variables, and these variables are often not observable in clinical environments. To close this gap, we propose the latent hybridisation model (LHM) that integrates a system of expert-designed ODEs with machine-learned Neural ODEs to fully describe the dynamics of the system and to link the expert and latent variables to observable quantities. We evaluated LHM on synthetic data as well as real-world intensive care data of COVID-19 patients. LHM consistently outperforms previous works, especially when few training samples are available such as at the beginning of the pandemic. © 2021 Neural information processing systems foundation. All rights reserved.

Recurrent Multi-task Graph Convolutional Networks for COVID-19 Knowledge Graph Link Prediction

Knowledge graphs (KGs) are a way to model data involving intricate relations between a number of entities. Understanding the information contained in KGs and predicting what hidden relations may be present can provide valuable domain-specific knowledge. Thus, we use data provided by the 5th Annual Oak Ridge National Laboratory Smoky Mountains Computational Sciences Data Challenge 2 as well as auxiliary textual data processed with natural language processing techniques to form and analyze a COVID-19 KG of biomedical concepts and research papers. Moreover, we propose a recurrent graph convolutional network model that predicts both the existence of novel links between concepts in this COVID-19 KG and the time at which the link will form. We demonstrate our model’s promising performance against several baseline models. The utilization of our work can give insights that are useful in COVID-19-related fields such as drug development and public health. All code for our paper is publicly available at https://github.com/RemingtonKim/SMCDC2021. © 2022, Springer Nature Switzerland AG.

Online, Interactive Tool for Studying How Students Troubleshoot Circuits

We have developed a new tool to look at how students interact with circuits during the troubleshooting process. The online tool was originally designed to analyze individual troubleshooting strategy for large classes, but it also works well in the COVID-era to facilitate remote learning. While there are a number of tools that allow students to virtually interact with circuits, none supported both breadboard graphics and recording all student interactions, which were necessary to create an authentic troubleshooting situation that could be analyzed by the researchers afterwards. Therefore, we created our own circuit and data analysis tool using HTML5, CSS, and JavaScript, which utilizes breadboard imagery from Fritzing and runs on most modern browsers. Unlike a traditional paper-and-pencil test, the interactive, online tool allows us to see how students react to new information and measure domain knowledge beyond theory-including interpreting physical circuits and making measurements. Instead of relying on students to tell us everything on their mind, we can use their actions as a proxy for their thought processes. This paper describes how we developed the tool and some preliminary data on how students debug. © American Society for Engineering Education, 2021

Transfer learning approaches to recognize X-ray Covid-19 images

Covid 19, caused from coronavirus SAR-CoV-2, is currently a dangerous threat to human beings. The rapid development of the Covid 19 pandemic forced all countries to develop fast and reliable methods to detect the coronavirus SAR-CoV-2. Transfer learning with medical images is a suitable such detecting method. Transfer learning, a deep learning technique, has special abilities such as speed of training, fewer requirements of training data set size and reduced demand of expert domain knowledge. Diagnosing Covid 19 using medical images is also considered by some to be more reliable than using traditional laboratory methods. This paper proposes transfer learning methods combined with medical images to detect Covid 19. Using a Covid 19 X-ray data set from Kaggle, this research considers viral pneumonia as a separate class, increasing the performance since viral pneumonia is often wrongly classified as Covid 19, even by radiologists. This paper uses specialized metrics to deal with the imbalanced nature of the data and visualises results using Local Interpretable Model-agnostic Explanations to indicate areas of images associated with Covid 19. The ResNet family of CNNs performed well, with ResNet 34 performing better than the 18 and 50 layer versions. Inception and DenseNet also have good classification performance. © 2021 IEEE.