Critical success and failure factors in the AI lifecycle: a knowledge graph-based ontological study

Journal of Modelling in Management, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 14, 2025

Purpose The purpose of this study is to provide a holistic understanding the factors that either promote or hinder adoption artificial intelligence (AI) in supply chain management (SCM) and operations (OM). By segmenting AI lifecycle examining interactions between critical success failure factors, aims offer predictive insights can help proactively managing these ultimately reducing risk failure, facilitating smoother transition into AI-enabled SCM OM. Design/methodology/approach This develops knowledge graph model lifecycle, divided pre-development, deployment post-development stages. methodology combines comprehensive literature review for ontology extraction expert surveys establish relationships among ontologies. Using exploratory factor analysis, composite reliability average variance extracted ensures validity constructed dimensions. Pearson correlation analysis applied quantify strength significance entities, providing metrics labeling edges resource description framework. Findings identifies 11 dimensions integration OM: (1) setting clear goals standards; (2) ensuring accountable with leadership-driven strategies; (3) activating leadership bridge expertise gaps; (4) gaining competitive edge through partnerships advanced IT infrastructure; (5) improving data quality customer demand; (6) overcoming resistance via awareness benefits; (7) linking domain infrastructure robustness; (8) enhancing stakeholder engagement effective communication; (9) strengthening robustness change training governance; (10) using key performance indicators-driven reviews management; (11) accountability copyright integrity governance. Originality/value enhances decision-making by developing segments stages, introducing novel approach OM research. incorporating element uses graphs anticipate outcomes from These assist practitioners making informed decisions about use, overall

Language: Английский

---

Prostate cancer management with lifestyle intervention: From knowledge graph to Chatbot

Clinical and Translational Discovery, Journal Year: 2022, Volume and Issue: 2(1)

Published: Feb. 20, 2022

Abstract Background Personal lifestyle is an important cause of prostate cancer (PCa), hence establishing a corresponding knowledge graph (KG) and chatbot convenient way for preventing assessing risks. The based on KG PCa‐associated lifestyles will be helpful to PCa management, then save health care resources in the ageing society. Results Based our established base, we define entities relationships construct visualization by importing triples into Neo4j server. dialogue system uses Flask framework determine classification questions through entity recognition relationship extraction later query template search answers from KG. contains 11 types 14 relationships, total number nodes links 21 546 66 493, respectively. Also, “Lifestyle”, “Paper”, “Baseline” “Outcome” contain multiple attributes. can answer 12 basic predict probability certain resulting PCa. available at http://sysbio.org.cn:5000/Pca/chatbot . Conclusion A was constructed help researchers, physicians or patients learn more about management interactively.

Language: Английский

---

Reliable knowledge graph fact prediction via reinforcement learning

Visual Computing for Industry Biomedicine and Art, Journal Year: 2023, Volume and Issue: 6(1)

Published: Nov. 20, 2023

Knowledge graph (KG) fact prediction aims to complete a KG by determining the truthfulness of predicted triples. Reinforcement learning (RL)-based approaches have been widely used for prediction. However, existing largely suffer from unreliable calculations on rule confidences owing limited number obtained reasoning paths, thereby resulting in decisions Hence, we propose new RL-based approach named EvoPath this study. features reward mechanism based entity heterogeneity, facilitating an agent obtain effective paths during random walks. also incorporates postwalking leverage easily overlooked but valuable RL. Both mechanisms provide sufficient facilitate reliable confidences, enabling make precise judgments about Experiments demonstrate that can achieve more accurate predictions than approaches.

Language: Английский

---

Comprehensive Personal Health Knowledge Graph for Effective Management and Utilization of Personal Health Data

Published: Feb. 5, 2024

The widespread use of electronic health records (EHRs) and wearable devices has generated a massive amount personal data (PHD) that can be utilized for research patient care. However, integrating managing various types PHD from different sources presents significant challenges, including interoperability, privacy, security concerns. In response, this paper proposes Personal Health Knowledge Graph integrated management utilization. This approach utilizes knowledge graphs to structure integrate sources, EHR data, device sensing insurance social determinants health. proposed offers comprehensive view an individual's health, allowing the integration analysis PHD. Additionally, three cases illustrate practical applications advantages (PHKG) in healthcare Overall, provides promising solution analyzing PHD, which used improve outcomes research.

Language: Английский

---

Leverage knowledge graph and GCN for fine-grained-level clickbait detection

World Wide Web, Journal Year: 2022, Volume and Issue: 25(3), P. 1243 - 1258

Published: March 16, 2022

Language: Английский

---

Insight Extraction From E-Health Bookings by Means of Hypergraph and Machine Learning

IEEE Journal of Biomedical and Health Informatics, Journal Year: 2023, Volume and Issue: 27(10), P. 4649 - 4659

Published: Jan. 2, 2023

New technologies are transforming medicine, and this revolution starts with data. Usually, health services within public healthcare systems accessed through a booking centre managed by local authorities controlled the regional government. In perspective, structuring e-health data Knowledge Graph (KG) approach can provide feasible method to quickly simply organize and/or retrieve new information. Starting from raw bookings system in Italy, KG is presented support extraction of medical knowledge novel insights. By exploiting graph embedding which arranges various attributes entities into same vector space, we able apply Machine Learning (ML) techniques embedded vectors. The findings suggest that KGs could be used assess patients' patterns, either unsupervised or supervised ML. particular, former determine possible presence hidden groups not immediately available original legacy dataset structure. latter, although performance algorithms very high, shows encouraging results predicting patient's likelihood undergo particular visit year. However, many technological advances remain made, especially database algorithms.

Language: Английский

---

A Survey of Advanced Information Fusion System: from Model-Driven to Knowledge-Enabled

Data Science and Engineering, Journal Year: 2023, Volume and Issue: 8(2), P. 85 - 97

Published: April 26, 2023

Abstract Advanced knowledge engineering (KE), represented by graph (KG), drives the development of various fields and technologies provides fusion empowerment interfaces. At same time, advanced system (SE) takes model-based (MBSE) as core to realize formal modeling process analysis whole system. The two complement each other are key for transition from 2.0 3.0 in era artificial intelligence perceptual cognitive intelligence. This survey summarizes an information system, model-driven knowledge-enabled. Firstly, concept, representative methods, application introduced. Then, it introduces concept knowledge-driven engineering, architecture construction methods fields. Finally, combination systems, opportunities challenges discussed.

Language: Английский

---

A Blockchain-Based Personal Health Knowledge Graph for Secure Integrated Health Data Management

2022 IEEE Symposium on Computers and Communications (ISCC), Journal Year: 2023, Volume and Issue: unknown, P. 1 - 7

Published: July 9, 2023

The increasing use of electronic health records (EHRs) and wearable devices has led to the creation massive amounts personal data (PHD) that can be utilized for research patient care. However, managing integrating various types PHD from different sources poses significant challenges, including interoperability, privacy, security. To address these this paper proposes a blockchain-based knowledge graph integrated management. proposed approach utilizes graphs structure integrate PHD, such as EHR, sensing, insurance data, provide comprehensive view an individual's health. blockchain ensure privacy By storing on decentralized platform, patients have full control over their grant access specific entities needed providing enhanced

Language: Английский

---

Zero-Shot Construction of Chinese Medical Knowledge Graph with GPT-3.5-turbo and GPT-4

ACM Transactions on Management Information Systems, Journal Year: 2024, Volume and Issue: unknown

Published: April 10, 2024

Knowledge graphs have revolutionized the organization and retrieval of real-world knowledge, prompting interest in automatic NLP-based approaches for extracting medical knowledge from texts. However, availability high-quality Chinese remains limited, posing challenges constructing graphs. As LLMs like ChatGPT show promise zero-shot learning many NLP downstream tasks, their potential on is still uncertain. In this study, we create a graph by manually annotating textual data using to automatically generate graph. We refine results filtering mapping rules align with our schema. The generated serves as ground truth evaluation, explore different methods enhance its accuracy through completion techniques. result, emphasize employing automated construction within domain. While successfully identifies larger number entities, further enhancements are required improve performance more qualified relations.

Language: Английский

---

Heterogeneous network-based algorithms in the biomedical data mining: A review from technical perspective

Informatics and Health, Journal Year: 2024, Volume and Issue: 1(2), P. 111 - 122

Published: Aug. 24, 2024

Language: Английский

---