Domain Knowledge Research Articles

A visual analysis approach for data transformation via domain knowledge and intelligent models

A visual analysis approach for data transformation via domain knowledge and intelligent models

Multi-granularity learning of explicit geometric constraint and contrast for label-efficient medical image segmentation and differentiable clinical function assessment

Automated segmentation is a challenging task in medical image analysis that usually requires a large amount of manually labeled data. However, most current supervised learning based algorithms suffer from insufficient manual annotations, posing a significant difficulty for accurate and robust segmentation. In addition, most current semi-supervised methods lack explicit representations of geometric structure and semantic information, restricting segmentation accuracy. In this work, we propose a hybrid framework to learn polygon vertices, region masks, and their boundaries in a weakly/semi-supervised manner that significantly advances geometric and semantic representations. Firstly, we propose multi-granularity learning of explicit geometric structure constraints via polygon vertices (PolyV) and pixel-wise region (PixelR) segmentation masks in a semi-supervised manner. Secondly, we propose eliminating boundary ambiguity by using an explicit contrastive objective to learn a discriminative feature space of boundary contours at the pixel level with limited annotations. Thirdly, we exploit the task-specific clinical domain knowledge to differentiate the clinical function assessment end-to-end. The ground truth of clinical function assessment, on the other hand, can serve as auxiliary weak supervision for PolyV and PixelR learning. We evaluate the proposed framework on two tasks, including optic disc (OD) and cup (OC) segmentation along with vertical cup-to-disc ratio (vCDR) estimation in fundus images; left ventricle (LV) segmentation at end-diastolic and end-systolic frames along with ejection fraction (LVEF) estimation in two-dimensional echocardiography images. Experiments on nine large-scale datasets of the two tasks under different label settings demonstrate our model’s superior performance on segmentation and clinical function assessment.

Preparing to teach relationships and sexuality education: student teachers’ perceived knowledge and attitudes

ABSTRACT Children have a right to Comprehensive Sexuality Education (CSE). A key place for this to occur is in schools, and teachers’ comfort and competence in teaching is critical to delivering high-quality CSE. Good quality Initial Teacher Education (ITE) should provide student teachers with a strong foundation for the delivery of CSE and the creation of safe, supportive, and inclusive classrooms and whole-school cultures. This study examined primary ITE students’ perceptions of the quality of input, knowledge and confidence, and attitudes towards Relationships and Sexuality Education (RSE). The study was conducted in Ireland where RSE is mandatory in the primary school curriculum. The level of input student teachers received in RSE was perceived to be insufficient, and less than that provided in other curricular areas. Student teachers reported moderate to high levels of perceived knowledge and comfort in RSE domains and high levels of commitment and intention to teach RSE once qualified. They overwhelmingly agreed that children and young people have the right to RSE. Student teachers reported being particularly unprepared to teach RSE to students with special educational needs and sexual and gender minority students. Recommendations are made for strengthening RSE input in ITE programmes.

Comparing the effects of patient safety education using design thinking and case based learning on nursing students’ competence and professional socialization: A quasi-experimental design

BackgroundPatient safety issues should be constantly monitored and sensitively recognized. In nursing education, it is necessary to find effective teaching methods to increase students' competencies in patient safety. ObjectivesThis study aimed to compare the effectiveness of a new method, design thinking (DT), and a traditional method, case-based learning (CBL), in patient safety education. DesignThis study used a quasi-experimental, pre-post control group design. SettingsA 30-h training tutorial was developed for intervention groups, DT and CBL, while the control group received no treatment. ParticipantsIn this study were junior nursing students receiving baccalaureate nursing education with experiences of at least 480-h clinical practice. A total of 53 students (21 in the DT group, 19 in the CBL group, and 13 in the control group) were recruited. MethodsClinical reasoning competency, patient safety competency, and professional socialization were measured immediately after the end of the educational program using DT and CBL, 4 weeks later, and 8 weeks later. Effects of DT and CBL were analyzed using a generalized estimating equation. ResultsBoth DT and CBL were effective in clinical reasoning competency (χ2 = 15.432, p = 0.017) and knowledge domain of patient safety competency (χ2 = 42.824, p < 0.001), showing no significant difference between the two. CBL was more effective in professional socialization than DT. ConclusionsDT was as effective as CBL in clinical reasoning competency and knowledge domain of patient safety competency. In the healthcare field, where improving the patient experience is becoming increasingly important, DT is worth applying as an educational method to train nursing students who can take a creative and human-centered problem-solving approach. It is expected that educational curricula utilizing DT will be developed not only in the field of patient safety but also in various patient care areas.

An interpretable multi-scaled agent hierarchy for time series prediction

The traditional time series analysis treats the time series as a dynamic system of sequential entries, leading to complex models and a lack of interpretability. Time series, however, can be understood simply as a composition of correlated information hierarchies in various time scales. Rising from this perspective, here we propose Interpretable multi-time-Scaled Hierarchical Agent-based Time Series Prediction (ISHA-TSP) with linear AutoRegressive (AR) architecture. These time scales, determined based on the instantaneous dominant frequencies of the time series or domain of knowledge, down-sample time series for each agent. Accordingly, a trend prediction signal is offered from a higher time-scale agent to a lower time-scale agent in a hierarchical manner. This hierarchy can also be interpreted reversely, where the lower time-scale agent offers a direction gradient signal to the higher-scale agent to adjust its estimated trend. The optimal learning rate for ISHA-TSP is determined based on a composite Lyapunov function that guarantees convergence. The proposed ISHA-TSP is evaluated with seven datasets in various prediction horizons for accuracy and stop-learning and stop-forecasting steps for robustness. Results show the ISHA-TSP is superior over an equivalent single agent in prediction error. Also, the proposed method adapts against agent failure and data-missing situations. This strategy also leads to simpler models with fewer parameters than a single complex monolithic model. Furthermore, it shows higher interpretability due to its compositional approach.

An analysis of the literature on construction employee turnover: drivers, consequences, and future direction

Employee turnover is a critical factor affecting organizational effectiveness. It is particularly important in the construction sector due to its knowledge and labor-intensive characteristics. Therefore, it is necessary to understand why qualified employees would leave their employers so that effective measures can be taken to retain them. Despite the growing body of knowledge on employee turnover, there is a lack of comprehensive review of this topic in the construction literature. This study adopts a systematic literature review approach to synthesize domain knowledge on employee turnover from two perspectives: its investigated causes, and consequences. In total, 77 relevant papers are extracted. A total of 51 drivers of construction employee turnover are identified and categorized into personal, job-related, organizational, and external influencers. Findings also suggest three major outcomes caused by employee turnover, including individual, project, and organization-related consequences. This review contributes to understanding the turnover phenomenon in construction literature and highlights potential areas for future empirical studies to move forward. By understanding the main factors influencing turnover decisions, construction organizations can develop and implement effective means for managing employee turnover. This study further contributes to construction employee turnover research by emphasizing the importance of studying the consequences of the turnover phenomenon.

Integrating Domain Knowledge with Deep Learning Model for Automated Worker Activity Classification in mobile work zone

Integrating Domain Knowledge with Deep Learning Model for Automated Worker Activity Classification in mobile work zone

Building a knowledge graph to enrich ChatGPT responses in manufacturing service discovery

Sourcing and identification of new manufacturing partners is crucial for manufacturing system integrators to enhance agility and reduce risk through supply chain diversification in the global economy. The advent of advanced large language models has captured significant interest, due to their ability to generate comprehensive and articulate responses across a wide range of knowledge domains. However, the system often falls short in accuracy and completeness when responding to domain-specific inquiries, particularly in areas like manufacturing service discovery. This research explores the potential of leveraging Knowledge Graphs in conjunction with ChatGPT to streamline the process for prospective clients in identifying small manufacturing enterprises. In this study, we propose a method that integrates bottom-up ontology with advanced machine learning models to develop a Manufacturing Service Knowledge Graph from an array of structured and unstructured data sources, including the digital footprints of small-scale manufacturers throughout North America. The Knowledge Graph and the learned graph embedding vectors are leveraged to tackle intricate queries within the digital supply chain network, responding with enhanced reliability and greater interpretability. The approach highlighted is scalable to millions of entities that can be distributed to form a global Manufacturing Service Knowledge Network Graph that can potentially interconnect multiple types of Knowledge Graphs that span industry sectors, geopolitical boundaries, and business domains. The dataset developed for this study, now publicly accessible, encompasses more than 13,000 manufacturers’ weblinks, manufacturing services, certifications, and location entity types.

A data set of Japanese corporates’ green transformation technologies patents

Abstract* In 2020, the Government of Japan declared 2050 “carbon neutral” and launched a long-term strategy to create a “virtuous cycle of economy and environment.” Japanese corporations possess many technologies that contribute to decarbonization, which is important for expanding investments in green transformation technology inventory (GXTI) development. Patent data are the most reliable measure of business performance for applied research and development activities when investigating knowledge domains or technological evolution. Our paper describes a Japanese patent dataset of Japanese corporations’ green transformation (GX) patent applications on the Japan Platform for Patent Information, using a search method of bombinating International Patent Classification (IPC) codes and keywords. The dataset contains 37,476 GX patent applications from 298 corporations during the period 1999–2022.

Novice Teachers’ Knowledge of Racial Literacies

ABSTRACT Grounded in Goodwin and Darity’s domains of knowledge for social justice educators, this study examines novice teachers’ (NTs) racial literacies. Specifically, this research integrates a framework for teacher knowledge with the construct of racial literacies to examine the structure of racially literate teachers’ knowledge and practices. Drawing on this framework, this research entailed designing, piloting, and validating a measure- the teachers’ Knowledge of Racial Literacies survey- of racial literacies. Semi-structured interviews of survey respondents were analyzed to improve the measure and reveal how participants’ racial literacies were developed. Findings indicate that NTs’ racial literacies exist on a continuum; participants demonstrated knowledge in individual domains, but not equally across domains. Additionally, nearly all participants reported few opportunities in teacher education programs to examine race and racism. Findings suggest the continued need for teacher education programs to develop connections between the five domains of knowledge to develop NTs as racially literate educators.

3DPFIX: Improving Remote Novices' 3D Printing Troubleshooting through Human-AI Collaboration Design

The widespread consumer-grade 3D printers and learning resources online enable novices to self-train in remote settings. While troubleshooting plays an essential part of 3D printing, the process remains challenging for many remote novices even with the help of well-developed online sources, such as online troubleshooting archives and online community help. We conducted a formative study with 76 active 3D printing users to learn how remote novices leverage online resources in troubleshooting and their challenges. We found that remote novices cannot fully utilize online resources. For example, the online archives statically provide general information, making it hard to search and relate their unique cases with existing descriptions. Online communities can potentially ease their struggles by providing more targeted suggestions, but a helper who can provide custom help is rather scarce, making it hard to obtain timely assistance. We propose 3DPFIX, an interactive 3D troubleshooting system powered by the pipeline to facilitate Human-AI Collaboration, designed to improve novices' 3D printing experiences and thus help them easily accumulate their domain knowledge. We built 3DPFIX that supports automated diagnosis and solution-seeking. 3DPFIX was built upon shared dialogues about failure cases from Q&amp;A discourses accumulated in online communities. We leverage social annotations (i.e., comments) to build an annotated failure image dataset for AI classifiers and extract a solution pool. Our summative study revealed that using 3DPFIX helped participants spend significantly less effort in diagnosing failures and finding a more accurate solution than relying on their common practice. We also found that 3DPFIX users learn about 3D printing domain-specific knowledge. We discuss the implications of leveraging community-driven data in developing future Human-AI Collaboration designs.

Is a Seat at the Table Enough? Engaging Teachers and Students in Dataset Specification for ML in Education

Despite the promises of ML in education, its adoption in the classroom has surfaced numerous issues regarding fairness, accountability, and transparency, as well as concerns about data privacy and student consent. A root cause of these issues is the lack of understanding of the complex dynamics of education, including teacher-student interactions, collaborative learning, and classroom environments. To overcome these challenges and fully utilize the potential of ML in education, software practitioners need to work closely with educators and students to fully understand the context of the data (the backbone of ML applications) and collaboratively define the ML data specifications. To gain a deeper understanding of such a collaborative process, we conduct ten co-design sessions with ML software practitioners, educators, and students. In the sessions, teachers and students work with ML engineers, UX designers, and legal practitioners to define dataset characteristics for a given ML application. We find that stakeholders contextualize data based on their domain and procedural knowledge, proactively design data requirements to mitigate downstream harms and data reliability concerns, and exhibit role-based collaborative strategies and contribution patterns. Further, we find that beyond a seat at the table, meaningful stakeholder participation in ML requires structured supports: defined processes for continuous iteration and co-evaluation, shared contextual data quality standards, and information scaffolds for both technical and non-technical stakeholders to traverse expertise boundaries.

An orchestra of machine learning methods reveals landmarks in single-cell data exemplified with aging fibroblasts.

In this work, a Python framework for characteristic feature extraction is developed and applied to gene expression data of human fibroblasts. Unlabeled feature selection objectively determines groups and minimal gene sets separating groups. ML explainability methods transform the features correlating with phenotypic differences into causal reasoning, supported by further pipeline and visualization tools, allowing user knowledge to boost causal reasoning. The purpose of the framework is to identify characteristic features that are causally related to phenotypic differences of single cells. The pipeline consists of several data science methods enriched with purposeful visualization of the intermediate results in order to check them systematically and infuse the domain knowledge about the investigated process. A specific focus is to extract a small but meaningful set of genes to facilitate causal reasoning for the phenotypic differences. One application could be drug target identification. For this purpose, the framework follows different steps: feature reduction (PFA), low dimensional embedding (UMAP), clustering ((H)DBSCAN), feature correlation (chi-square, mutual information), ML validation and explainability (SHAP, tree explainer). The pipeline is validated by identifying and correctly separating signature genes associated with aging in fibroblasts from single-cell gene expression measurements: PLK3, polo-like protein kinase 3; CCDC88A, Coiled-Coil Domain Containing 88A; STAT3, signal transducer and activator of transcription-3; ZNF7, Zinc Finger Protein 7; SLC24A2, solute carrier family 24 member 2 and lncRNA RP11-372K14.2. The code for the preprocessing step can be found in the GitHub repository https://github.com/AC-PHD/NoLabelPFA, along with the characteristic feature extraction https://github.com/LauritzR/characteristic-feature-extraction.

Knowledge-driven deep learning for fast MR imaging: Undersampled MR image reconstruction from supervised to un-supervised learning.

Deep learning (DL) has emerged as a leading approach in accelerating MRI. It employs deep neural networks to extract knowledge from available datasets and then applies the trained networks to reconstruct accurate images from limited measurements. Unlike natural image restoration problems, MRI involves physics-based imaging processes, unique data properties, and diverse imaging tasks. This domain knowledge needs to be integrated with data-driven approaches. Our review will introduce the significant challenges faced by such knowledge-driven DL approaches in the context of fast MRI along with several notable solutions, which include learning neural networks and addressing different imaging application scenarios. The traits and trends of these techniques have also been given which have shifted from supervised learning to semi-supervised learning, and finally, to unsupervised learning methods. In addition, MR vendors' choices of DL reconstruction have been provided along with some discussions on open questions and future directions, which are critical for the reliable imaging systems.

Customer Churn Prediction

In today’s highly competitive business landscape, customer retention revenue preservation, customer experience improvement, and marketing optimisation are critical factors for sustained growth and profitability. Customer churn predictionis discontinuing their services or purchases, which presents asignificant challenge for businesses across various industries. This project focuses on developing a predictive model to expect customer churn in the banking sector using machine learning techniques. The project involves the collection and analysis of historical customer data, confined account activity, transaction history, demographic information, and customer service interactions. By extracting the right features from this data, a machine learning model is trained to forecast which bank customers are at the highest risk of churning. A critical step in this study was the selection of relevant features that influence customerchurn. Feature selection was guided by domain knowledge and feature importance analysis. The different classifiers were used and then trained on the training dataset further ensuringthe model’s optimal performance. The model’s performance is assessed through various evaluation metrics, including accuracy, precision, and recall. Additionally, the project explores a model illustration to uncover the influential factors contributing to customer churn within the banking context. This project’s outcomes can empower banks to take proactive measures in retaining customers, enhancing their overall experience, and thereby preserving revenue streams. By addressing customer churn, banks can foster long-term relationships, reduce customer acquisition costs, and boost their competitiveness in the financial industry. The results of this project are expected to assist businesses in proactively retaining customers by targeting those at the highest risk of churning. Ultimately, reducing customer churn can lead to increased customer satisfaction, revenue, and long-term business sustainability.

IKDSumm: Incorporating key-phrases into BERT for extractive disaster tweet summarization

Online social media platforms, such as Twitter, are one of the most valuable sources of information during disaster events. Humanitarian organizations, government agencies, and volunteers rely on a concise compilation of such information for effective disaster management. Existing methods to make such compilations are mostly generic summarization approaches that do not exploit domain knowledge. In this paper, we propose a disaster-specific tweet summarization framework, IKDSumm, which initially identifies the crucial and important information from each tweet related to a disaster through key-phrases of that tweet. We identify these key-phrases by utilizing the domain knowledge (using existing ontology) of disasters without any human intervention. Further, we utilize these key-phrases to automatically generate a summary of the tweets. Therefore, given tweets related to a disaster, IKDSumm ensures fulfillment of the summarization key objectives, such as information coverage, relevance, and diversity in summary without any human intervention. We evaluate the performance of IKDSumm with 8 state-of-the-art techniques on 12 disaster datasets. The evaluation results show that IKDSumm outperforms existing techniques by approximately 2−79% in terms of ROUGE-N F1-score.

Discovering urban block typologies in Seoul: Combining planning knowledge and unsupervised machine learning

Urban blocks are fundamental elements of urban form and play an important role in urban development. Identification of their typologies has been a long-standing challenge in urban planning and design. Traditionally, expert-based classification has been used, but recent studies have shifted towards data-driven clustering to achieve better results. However, most studies tend to use only one of these approaches and few integrate both. To address this gap, this study proposes a framework that combines planning knowledge and machine learning to identify urban block typologies in complex urban environments. The framework first classifies blocks based on their size types according to planning regulations and development types. Then, a combined hierarchical and Gaussian mixture model is used to discover clusters with complex feature data distributions. These clusters are translated into block typologies using a semantic naming system and illustrated with representative blocks. Finally, the validity of the results is tested by comparing them with other building features, such as building age. The proposed integrated framework was applied to Seoul, resulting in the discovery of 27 residential block typologies across the entire city. These typologies show significant correlations with the average building age in the blocks. The application of expert knowledge in the clustering method demonstrates higher accuracy compared to conventional clustering approaches. Overall, this framework holds significant potential for discovering urban block typologies that are consistent with planning domain knowledge and provide urban planners and designers with valuable insights into the characteristics of urban block typologies, thereby helping decision-making regarding zoning regulations, urban design, and urban development strategies.

Validating the knowledge represented by a self-organizing map with an expert-derived knowledge structure.

Professionals are reluctant to make use of machine learning results for tasks like curriculum development if they do not understand how the results were generated and what they mean. Visualizations of peer reviewed medical literature can summarize enormous amounts of information but are difficult to interpret. This article reports the validation of the meaning of a self-organizing map derived from the Medline/PubMed index of peer reviewed medical literature by its capacity to coherently summarize the references of a core psychiatric textbook. Reference lists from ten editions of Kaplan and Sadock's Comprehensive Textbook of Psychiatry were projected onto a self-organizing map trained on Medical Subject Headings annotating the complete set of peer reviewed medical research articles indexed in the Medline/PubMed database (MedSOM). K-means clustering was applied to references from every edition to examine the ability of the self-organizing map to coherently summarize the knowledge contained within the textbook. MedSOM coherently clustered references into six psychiatric knowledge domains across teneditions (1967-2017). Clustering occurred at the abstract level of broad psychiatric practice including General/adult psychiatry, Child psychiatry, and Administrative psychiatry. The uptake of visualizations of published medical literature by medical experts for purposes like curriculum development depends upon validation of the meaning of the visualizations. The current research demonstrates that a self-organizing map (MedSOM) can validate the stability and coherence of the references used to support the knowledge claims of a standard psychiatric textbook, linking the products of machine learning to a widely accepted standard of knowledge.

How generative AI will drive enterprise innovation

PurposeMost recent C-suite surveying suggests current applications of generative AI, although hyped, are fragmented and unlikely to yield major financial returns anticipated. Instead, business leaders expect major value from generative AI will be achieved through application of generative AI to innovation: operational innovation, product and service innovation, and most elusive of all, business model innovation.Design/methodology/approachFindings and analysis presented draws on data from several surveys of C-level executives conducted by IBM Institute for Business Value in collaboration with Oxford Economics during 2023. Each survey focused on the potential of generative AI in a particular business area. The n-count of each survey ranged from 100-3000.Findings1. Business leaders expect generative AI to build on returns achieved from investments in traditional AI, with 10 percent RoI expected on generative AI investments by 2025. 2. Executives anticipate that generative AI will have most impact when implemented to expand innovation. 3. Specific examples provided for operational innovation, product innovation, and business model innovationResearch limitations/implicationsWe are still very early in the generative AI development cycle. We have made best efforts to project, but only time will tell for sure.Practical implicationsBusiness application of generative AI are extremely fragmented. Despite the desire to throw investments at the wall to see what sticks, it is important that leaders take a structured approach to generative AI, focusing on RoI from innovation investments.Social implicationsTo alleviate negative impacts of generative AI, focusing on innovation potential and value maximization is crucial.Originality/valueThis research is based on completely new surveying and data. This papers adds to the sum total of new knowledge in the generative AI domain.

HaLo‐NeRF: Learning Geometry‐Guided Semantics for Exploring Unconstrained Photo Collections

AbstractInternet image collections containing photos captured by crowds of photographers show promise for enabling digital exploration of large‐scale tourist landmarks. However, prior works focus primarily on geometric reconstruction and visualization, neglecting the key role of language in providing a semantic interface for navigation and fine‐grained understanding. In more constrained 3D domains, recent methods have leveraged modern vision‐and‐language models as a strong prior of 2D visual semantics. While these models display an excellent understanding of broad visual semantics, they struggle with unconstrained photo collections depicting such tourist landmarks, as they lack expert knowledge of the architectural domain and fail to exploit the geometric consistency of images capturing multiple views of such scenes. In this work, we present a localization system that connects neural representations of scenes depicting large‐scale landmarks with text describing a semantic region within the scene, by harnessing the power of SOTA vision‐and‐language models with adaptations for understanding landmark scene semantics. To bolster such models with fine‐grained knowledge, we leverage large‐scale Internet data containing images of similar landmarks along with weakly‐related textual information. Our approach is built upon the premise that images physically grounded in space can provide a powerful supervision signal for localizing new concepts, whose semantics may be unlocked from Internet textual metadata with large language models. We use correspondences between views of scenes to bootstrap spatial understanding of these semantics, providing guidance for 3D‐compatible segmentation that ultimately lifts to a volumetric scene representation. To evaluate our method, we present a new benchmark dataset containing large‐scale scenes with ground‐truth segmentations for multiple semantic concepts. Our results show that HaLo‐NeRF can accurately localize a variety of semantic concepts related to architectural landmarks, surpassing the results of other 3D models as well as strong 2D segmentation baselines. Our code and data are publicly available at https://tau‐vailab.github.io/HaLo‐NeRF/.