Knowledge Reasoning Research Articles

MvTuckER: Multi-view knowledge graphs representation learning based on tensor tucker model

Recently, more and more multi-view knowledge graphs with various important attributes are being constructed and applied, such as temporal information, geolocation, commonsense knowledge, multi-language and multi-modal. Knowledge graphs (KGs) have been widely applied in question answering systems, drug discovery, information retrieval, and so on. Knowledge graph representation learning (KGRL) is emerging as the most effective approach to tasks such as knowledge graph completion, entity alignment, knowledge reasoning and knowledge query. However, most previous work on KGs and KGRL are mainly designed for traditional KGs and only consider single-view, and the KGRL methods for special multi-view KGs faces the challenges of lacking a unified framework, insufficient interactive information and underutilizing important attributes. To bridge this gap, in this paper, we proposes a multi-view knowledge graphs (e.g., temporal KGs) representation learning method based on tensor tucker model (MvTuckER), which models the multi-view KGs from tuples to a nth-order binary tensor, and adopts the tensor n-mode product to capture complete interaction between and within different views. We also introduce low-rank and sparse approximations of core tensors to balance expressivity and complexity of MvTuckER. Moreover, we theoretically illustrate that our model is fully expressive for modeling multi-view KGs, discuss the working mechanism of MvTuckER from the perspective of logical operation, and explain how to migrate and extend our method to other types of views. We conducted extensive experiments on three multi-view knowledge graphs, and obtain 4.7%, 2.7%, and 5.3% effective improvement in Hit@1 respectively, which proves that the proposed MvTuckER can achieve state-of-the-art performance.

Leveraging Conflict to Bridge Cognitive Reasoning and Generative Algorithms

Autonomous agents require the ability to identify and adapt to unexpected conditions. Real-world environments are rarely stationary, making it problematic for agents operating in such environments to learn efficient policies. There is therefore a need for a general framework capable of detecting when an agent has encountered novel conditions, and determining how it should adjust its actions. In this position paper we propose a framework that couples cognitive reasoning and generative algorithms by leveraging conflict detection to adjust to unexpected dynamics. Specifically, we propose that a metacognitive conflict resolution mechanism is necessary; such a mechanism would balance the use of commonsense and deliberative reasoning to allow the agent to navigate novel conditions.

Knowledge reasoning in power grid infrastructure projects based on deep multi-view graph convolutional network

With the rapid development of power grid infrastructure, especially the increasing number of ultra-high voltage (UHV) projects, knowledge extracted from historical engineering data is collected and can be potentially used to assist in the review of power transmission and transformation projects. However, conventional knowledge modeling and knowledge reasoning methods cannot meet the current needs of power grid construction. In this paper, considering the more supernumerary and distinctive information brought by multi-view data which could be beneficial for feature representation and knowledge reasoning from the constructed knowledge base, a multi-view graph convolutional network (GCN) based on knowledge graph is proposed to make classification for power grid infrastructure projects. Specifically, several views are constructed based on attribute information of a knowledge graph. In addition, a Haar convolution-based pooling mechanism is employed to capture the structural features represented by a chain of subgraphs. And then an aggregator that combines both attribute and structural information is used to classify UHV projects. Results from both UHV and NCI-1 datasets indicate that our proposed method is more has higher accuracy and generalization ability.

Convolutional Dimension-Reduction With Knowledge Reasoning for Reliability Approximations of Structures Under High-Dimensional Spatial Uncertainties

Abstract Along with the rapid advancement of additive manufacturing technology, 3D-printed structures and materials have been successfully employed in various applications. Computer simulations of these structures and materials are often characterized by a vast number of spatial-varied parameters to predict the structural response of interest. Direct Monte Carlo methods are infeasible for uncertainty quantification and reliability assessment of such systems as they require a large number of forward model evaluations to obtain convergent statistics. To alleviate this difficulty, this paper presents a convolutional dimension-reduction method with knowledge reasoning-based loss regularization for surrogate modeling and uncertainty quantification of structures with high-dimensional spatial uncertainties. To manage the inherent high-dimensionality, a deep convolutional dimension-reduction network (ConvDR) is constructed to transform the spatial data into a low-dimensional latent space. In the latent space, knowledge reasoning is formulated as a form of loss regularization, and evolutionary algorithms are employed to train both the ConvDR network and a linear regression model as surrogate models for predicting the response of interest. 2D structures with spatial-variated material compositions are used to demonstrate the performance of the proposed approach.

The influence of fears of perceived legal consequences on general practitioners’ practice in relation to defensive medicine – a cross-sectional survey in Germany

BackgroundMedical decisions are influenced by a variety of factors also by legal requirements and feelings of uncertainty, which results in the term defensive medicine. The aim of the study was to evaluate the influence of fears of perceived legal consequences on the practice of defensive medicine from the perspective of German general practitioners (GPs).MethodsA cross-sectional study was performed from April to May 2022. GPs were invited via an e-mail newsletter of the Institute for Continuing Education in Family Medicine in the German Association of General Practitioners and via an online platform of the German College of General Practitioners and Family Physicians. The evaluation of legal fears, the general assessment of defensive medicine and reasons for and the frequency of defensive medical measures were surveyed in this study. Beside descriptive analyses, a stepwise linear regression analysis was used to explore potential associations between for the primary outcome variable ‘fears of legal consequences’ on the practice of defensive medicine.Results413 general practitioners with an average age of 50 years (51% female) responded. The majority rated their fears of legal consequences as low to average whereas for almost a third (27%, n = 113) the fears were strong to very strong. Regarding legal fears, the physician-patient-relationship played a fairly to very large role for 48% (n = 198) of the respondents. One third estimated the probability of being sued civilly in the next 10 years as rather high to very high. 47% (n = 193) of the participants assumed that the risk of being sued could mostly to very much be reduced by defensive medicine. Legal self-protection was for 38% of the responders (n = 157) quite frequently to very frequently a reason for acting defensively. Consequently, half of the respondents stated that they performed unnecessary laboratory tests at least once per week and 40% indicated that they referred patients for radiological diagnostics without medical indication once per month.ConclusionsAs legal fears have an influence on medical practice and legal self-protection being a frequent reason for defensive behaviour, understanding and knowledge of the law should be improved by legal education at university and further training of post-graduate trainees and practicing physicians should be implemented. Additionally, a more in-depth enlightenment of society about the phenomenon of Protective and Defensive Medicine and its consequences could be a possibility to decrease the perceived fears of legal consequences on the physicians’ side.

Feature aggregation-based multi-relational knowledge reasoning for COPD intelligent diagnosis

The increasing prevalence of artificial intelligence-based knowledge reasoning has contributed to more accurate and efficient auxiliary diagnoses. However, a majority of the disease prediction methods concentrate on the symptoms themselves while discarding the inherent properties of symptoms and the relationships underlying them. This paper proposes a feature aggregation-based intelligent diagnosis model employing a Heterogeneous Graph Convolutional Network (GCN), termed HeteroGCN. It focuses on symptoms’ inherent properties and multiple hidden relationships among symptoms and properties. By aggregating features of nodes, it realizes effective and accurate symptom-based knowledge reasoning for disease-type prediction. The diagnosis-related information from the Electronic Medical Record (EMR) has been extracted and standardized by taking chronic obstructive pulmonary disease (COPD) as an instance. Then the presented model extracts the symptoms and their properties as nodes and the relationships underlying the nodes as edges to construct a heterogeneous graph. The adjacency matrix and feature matrix have been fused and taken as the input of this model, and then the node representations (embeddings) are generated by aggregating neighbor nodes’ information. Finally, specific disease types (syndromes) will be predicted by the generated symptom node embeddings. The results of the model comparison and parameter sensitivity test demonstrate that the presented HeteroGCN model performs best on disease-type prediction. This paper provides a novel feature aggregation-based multi-relational knowledge reasoning approach for disease type (syndrome) prediction, which holds great significance in improving disease diagnosis.

Commonsense reasoning and automatic generation of IoT contextual knowledge: An Answer Set Programming approach

The increasing interest in searching for intelligent, proactive, and autonomous environments leads to the necessity of accessing contextual knowledge: knowledge about changes that occur in the environment and even about the capabilities of the devices that compose a specific deployment. This knowledge would allow carrying out commonsense reasoning (exhibit human-like comprehension) according to the current situation, enabling decision making and task planning. One of the most used forms of knowledge specification is ontologies, but ontological knowledge modeling is usually a manual process, making it a costly task. In addition, ontologies can be found that have been designed for specifying knowledge about IoT-related concepts, but many are overly complex or do not have an adequate orientation that allows modeling the system’s capabilities. There are many different types of reasoners, but Answer Set Solvers are of particular interest for commonsense reasoning. This paper proposes an ontology to represent contextual knowledge about IoT device capabilities, and an Answer Set Programming-based reasoner for the automatic generation of this knowledge. The main challenge is to demonstrate that contextual knowledge can be generated through commonsense reasoning processes implemented with Answer Set Programming, and that the resulting knowledge can be used for decision making (also through commonsense reasoning). This work shows how the generated knowledge is correct through different use cases, presents an application example that demonstrates the benefits of using the generated knowledge, and analyzes the reasoner performance to demonstrate that the execution time is adequate.

Semantically triggered qualitative simulation of a geological process

The field of geology has been the subject of a range of research efforts aiming to formalize geological domain knowledge, notably through geological domain ontologies. The main focus of existing geological ontologies primarily lies in describing static geological objects and their properties, paying less attention to the knowledge concerning geological processes and events. Meanwhile, the geological process modeling and simulation predominantly rely on quantitative numerical approaches that necessitate comprehensive and abundant data as input. However, many geological processes took place on a million-year time scale with insufficient data and non-direct observations. Given the inherent incompleteness of geological data, geologists still rely on qualitative reasoning to validate their interpretations. There is currently a dearth of applicable methods to facilitate qualitative reasoning and simulate geological processes based on domain knowledge.We propose to model the effects of a geological process through an object-oriented program, while keeping an ontological representation of the situation at each instant. To combine the two models, we propose using semantically defined ‘process triggers.’ These process triggers are defined as part of the ontology, in accordance with the Basic Formal Ontology. They enable geologists to describe the precise moment when a geological process is triggered and initiated. On the computational program side, we employ the ‘Semantic Micro Object Language’ to embody the knowledge and rules provided by geologists, facilitating the simulation of geological processes. Through an evaluation experiment, our proposed approach demonstrates promising results within a reasonable timeframe. As proof of concept, we have applied our method to a real-world scenario of petroleum thermal maturation in Ekofisk Field and got a promising result. Our approach provides a formalism that allows a powerful code to interact with domain ontologies, which paves the path for future knowledge reasoning.

IndRT-GCNets: Knowledge Reasoning with Independent Recurrent Temporal Graph Convolutional Representations

IndRT-GCNets: Knowledge Reasoning with Independent Recurrent Temporal Graph Convolutional Representations

Explainable AI for Medical Image Analysis in Medical Cyber-Physical Systems: Enhancing Transparency and Trustworthiness of IoMT.

Medical image analysis plays a crucial role in healthcare systems of Internet of Medical Things (IoMT), aiding in the diagnosis, treatment planning, and monitoring of various diseases. With the increasing adoption of artificial intelligence (AI) techniques in medical image analysis, there is a growing need for transparency and trustworthiness in decision-making. This study explores the application of explainable AI (XAI) in the context of medical image analysis within medical cyber-physical systems (MCPS) to enhance transparency and trustworthiness. To this end, this study proposes an explainable framework that integrates machine learning and knowledge reasoning. The explainability of the model is realized when the framework evolution target feature results and reasoning results are the same and are relatively reliable. However, using these technologies also presents new challenges, including the need to ensure the security and privacy of patient data from IoMT. Therefore, attack detection is an essential aspect of MCPS security. For the MCPS model with only sensor attacks, the necessary and sufficient conditions for detecting attacks are given based on the definition of sparse observability. The corresponding attack detector and state estimator are designed by assuming that some IoMT sensors are under protection. It is expounded that the IoMT sensors under protection play an important role in improving the efficiency of attack detection and state estimation. The experimental results show that the XAI in the context of medical image analysis within MCPS improves the accuracy of lesion classification, effectively removes low-quality medical images, and realizes the explainability of recognition results. This helps doctors understand the logic of the system's decision-making and can choose whether to trust the results based on the explanation given by the framework.

GATiT: An Intelligent Diagnosis Model Based on Graph Attention Network Incorporating Text Representation in Knowledge Reasoning

GATiT: An Intelligent Diagnosis Model Based on Graph Attention Network Incorporating Text Representation in Knowledge Reasoning

A knowledge-driven framework and methodology for precision assembly process decision making

Precision assembly process planning technology is one of the major challenges in the field of assembly. Compared with general assembly process planning, precision assembly process planning needs to consider a more complex assembly process, higher precision requirements, and more difficult process control. For this reason, it is of great significance to implement an intelligent assembly process planning and decision-making method for the development of precision assembly technology. This paper realizes the intelligent planning of precision assembly processes through the decision-making route of knowledge modeling and knowledge reasoning. Firstly, a theoretical system of a knowledge-driven precision assembly process decision-making framework is proposed. Secondly, a key precision assembly process decision-making technology is described in detail, which introduces a new approach for knowledge updates based on matching rules. Finally, the effectiveness of the knowledge update method is demonstrated through experimental cases. The knowledge-driven decision-making method combines the accuracy of domain experts’ empirical knowledge with the efficiency of intelligent reasoning algorithms. The knowledge update method ensures the currency of knowledge used for decision making. Therefore, the proposed precision assembly decision-making system framework and knowledge updating techniques in this paper are of great significance for knowledge-driven precision assembly process planning.

Knowledge graph revision in the context of unknown knowledge.

The role of knowledge graph encompasses the representation, organization, retrieval, reasoning, and application of knowledge, providing a rich and robust cognitive foundation for artificial intelligence systems and applications. When we learn new things, find out that some old information was wrong, see changes and progress happening, and adopt new technology standards, we need to update knowledge graphs. However, in some environments, the initial knowledge cannot be known. For example, we cannot have access to the full code of a software, even if we purchased it. In such circumstances, is there a way to update a knowledge graph without prior knowledge? In this paper, We are investigating whether there is a method for this situation within the framework of Dalal revision operators. We first proved that finding the optimal solution in this environment is a strongly NP-complete problem. For this purpose, we proposed two algorithms: Flaccid_search and Tight_search, which have different conditions, and we have proved that both algorithms can find the desired results.

Early diagnosis of Alzheime`s disease based on VGG19 feature extractor and MRI analysis

Alzheimer’s Disease (AD) is the most prevalent cause of dementia, a neurodegenerative\ndisorder primarily affecting the brain. Mild cases of dementia gradually worsen, resulting in a\ndecline in cognitive abilities, including memory, thinking, reasoning, and behavior. Currently,\nthere is no cure for AD; however, early diagnosis of mild cases can help slow the disease\'s\nprogression, improve its management, and enhance overall quality of life. Existing models for\nearly AD diagnosis encounter challenges, such as low accuracy and precision, poor\ngeneralization and overfitting due to data imbalances in datasets. To address these limitations, we\npropose a novel Deep Learning (DL) based model for early AD diagnosis. Our model comprises\nthree main components: image processing and data augmentation, feature extraction based on\nVGG19 (Visual Geometry Group 19), and a Multi-Layer Perceptron (MLP) based classifier\ncategorizing input Magnetic Resonance Imaging (MRI) scans into four classes representing AD\'s\ndevelopmental stages: Very-Mild-Demented, Mild-Demented, Moderate-Demented, and NonDemented. We optimized our model using the ADAM Algorithm and applied it to images selected\nfrom the Open Access Series of Imaging Studies (OASIS) dataset. Our model yielded impressive\nresults, achieving an accuracy of 92.58%, an AUC of 99%, and precision, recall, and F-measure\nvalues of 0.907, 0.894, and 0.890, respectively. Furthermore, we conducted a comparative\nanalysis with several state-of-the-art models, consistently demonstrating that our proposed model\noutperforms them.

Cognitive Constructs and Personality Profiles in Youth with Deliberate Self-Harm Behaviours: A Scoping Review

This scoping review explored the relationship between cognitive constructs and personality profiles in deliberate selfharm among youth, with the objective of proposing preventive and intervention strategies based on the findings. The review analysed the correlation between cognitive constructs and Deliberate Self-Harm behaviors in youth with diverse personality profiles. It investigated the role of cognitive distortions, emotional regulation, reasoning, and executive functioning in youth engaged in deliberate self-harm. A comprehensive search using the Scopus database yielded 72 relevant articles for analysis. The prevalence rates of deliberate self-harm among youth range from 7.5% to 46.5%, emphasizing the need for targeted interventions. Promising approaches include cognitive restructuring, mindfulnessbased therapies, and interventions focusing on challenging cognitive distortions to promote adaptive thinking patterns. Training interventions addressing executive functioning skills can effectively target impulsive and risky behaviors. Personality traits such as neuroticism, sensation seeking, borderline personality disorder, and perfectionism require interventions focused on emotional resilience, coping strategies, risk assessment, self-control, and impulse management

"Meeple-Centred Design" to Assess Collaborative Play: The Case of Team3

The social model of disability refers to the fact that disabilities occur when the surrounding environment isn’t prepared to accommodate the needs of People with Disabilities. Since accessibility has been given more importance in several areas of social life, we argue here that board games can also contribute to promoting the social life of People with Disabilities, since they can help combat their social isolation. As such, we will be doing an analysis of a game based on its accessibility. In this paper, we analyze the accessibility of the game Team3 in its two versions (green and pink), based on the guidelines ‘Meeple Centred Design’. Team3 is a cooperative board game developed to improve players’ collaboration, communication, and cognitive spatial reasoning abilities. The game is designed for three or more players, with each player taking on one of three roles: the architect, the builder, or the supplier. Upon examining both versions and their distinct challenges, we reached the conclusion that the game is accessible from both a physical and socioeconomic standpoint. We identified that the category related to intersectionality is where most issues arise. Furthermore, the game should come with a disclaimer saying that it can be challenging for neurodiverse people, people with the same sensory disability playing together, and people with multiple disabilities so that the players or the person that supports them can decide if the game is adequate for its intended purpose.

The Effect of Uncertainty Training on the Improvement of Diagnostic Ability in Chinese Medical Students.

To evaluate the effect of the uncertainty training on improvement of students' diagnostic ability. Data were collected on 70 fifth-year medical students enrolled in the Case Discussion courses on Obstetrics and Gynecology in the spring of 2020. Of these students, 36 were in the uncertainty training group and 34 in the control group. The effect of training was evaluated by cognitively diagnostic assessment which mapped exam questions to 4 attributes assessing clinical reasoning and basic science knowledge. Uncertainty training was able to improve students' ability to use basic science concepts for inference and problem solving, and the ability to integrate complex clinical information to arrive at a diagnosis. But it could not improve students' ability on the basic recall of foundational concepts and the ability to use basic science concepts in clinical reasoning. Medical students could do well in integrating complex clinical information although they didn't recall basic science knowledge well. Uncertainty training could be used as an effective teaching method in Case Discussion course on Obstetrics and Gynecology. However, students still need to improve their basic knowledge besides the training.

Development of AKM-Based Mathematics Questions in the Adiwiyata Context to Train Students' Numeracy Skills

This study aims to produce AKM-based mathematics questions in the Adiwiyata context that are valid and practical and potentially affect the numeracy abilities of class X and XI MA students. The research method used is a development studies type design, which consists of two stages: the preliminary and formative evaluation. Data collection techniques used were walk-throughs, interviews, questionnaires and tests. The validity of the questions is obtained from the results of the validator's assessment in terms of content, construct, and language at the expert review stage and the clarity and readability of the questions based on students' comments and suggestions at the one-to-one stage. The practicality of the questions is obtained at the small group stage based on the questionnaire results. The potential effect of questions is obtained at the field test stage based on test and questionnaire results. This research produced 30 mathematics questions based on AKM contexts that were valid and practical and potentially affected numeracy abilities. The results of testing mathematics questions based on AKM in the Adiwiyata context showed that 44.87% of the 156 (26 students x 6 questions) analyzed answer items were identified as correct answers. As many as 44.23% of the 52 subjects (26 students x 2 questions) could solve knowing-level questions. As many as 48.7% of 78 subjects (26 students x 3 questions) could solve questions at the applying cognitive level, and 42.3% of 26 (26 students x 1 question) subjects could solve questions at the cognitive reasoning level.

Justifying decision making in socio-scientific issues: the roles of reasoning and knowledge

ABSTRACT In the recent decade, the proliferation of ICTs that require Wi-Fi routers in schools has been accompanied by public concerns about risks fueled in many cases by media reporting. The current study examines ways in which parents of school-age children perceive the issue of Wi-Fi radiation in schools by using a science literacy framework that examined engagement with Wi-Fi as a socio-scientific issue. In this messy epistemological space, where parents are faced with too much information of uneven quality, the science literacy literature expects them to reach an informed decision. This mixed-method study includes interviews (study I, n = 35) and questionnaires (study II, n = 513) with parents to explore their attitudes, decisions, and justifications regarding questions about Wi-Fi in schools. The interviews demonstrated heuristics and shortcuts used for decision-making. Considerations often did not include in-depth investigation or scientific knowledge. The questionnaire further examined the patterns identified in the interviews and how they related to socio-demographics and scientific background. Findings indicate that neither scientific reasoning nor relevant knowledge was necessarily connected with the decisions made or the justification given by the parents. This should be of interest to future educational programs focusing on promoting science literacy for informed citizenship.

Parents' Knowledge and Skills in Feeding, Children's Response to Feeding on the Nutritional Status of Toddlers with Malnutrition Problems

Malnutrition, notably stunting, significantly impacts children's health and development, particularly during the first 1000 days of life. The prevalence of stunting in Indonesia is notably high, affecting approximately 30% of children under five in 2019, with Kediri Regency reporting a prevalence rate of 10.23% in 2022. The aim of this research is to analyze the influence of parental knowledge and skills in improving nutritional status as well as children's responses to feeding on the nutritional status of children with malnutrition problems. The research design used in this research is a pre-experimental one group pre-test post-test design. The sampling technique was total sampling with respondents in this study being 22 underprivileged families with toddlers who had nutritional problems and stunting in the Gempolan Village area. The problem of malnutrition in toddlers is quite high with an average Z Score of -2.42 (BB/U), -2.46 (TB/U), and -1.53 (BB/TB). Pre-test data on parents' knowledge and ability to fulfill children's nutrition shows that 68% have fairly good knowledge and skills and only 9% have good knowledge and skills. In this study, 50% of respondents had a fairly good response to feeding. The data highlights a significant problem of malnutrition in toddlers, characterized by low Z-Scores indicating undernutrition and growth stunting. These low Z-Scores pose substantial health risks, including compromised growth, weakened immunity, and potential long-term health issues. Parental knowledge and skills in nutrition are crucial, with most parents having reasonable knowledge levels. However, feeding problems persist, contributing to food rejection by children, exacerbating malnutrition.