Article
Chemistry, Physical
Yang Gao, Ludi Wang, Xueqing Chen, Yi Du, Bin Wang
Summary: Electrocatalysis plays a crucial role in sustainable fuel and chemical production. The combination of artificial intelligence and catalytic science shows great potential for extracting, analyzing, and predicting electrocatalysts. By constructing a knowledge graph based on a linguistically enriched SciBERT-based framework, the study focuses on Cu-based electrocatalysts for electrocatalytic CO2 reduction. The graph retrieves various entities from scientific literature, such as material, regulation method, product, and Faradaic efficiency, providing researchers with valuable insights.
Article
Computer Science, Interdisciplinary Applications
J. Raphael Seidenberg, Ahmad A. Khan, Alexei A. Lapkin
Summary: Conceptual process design involves finding optimal process flowsheets in a large design space. Effective approaches often rely on restricting the search space, which can be done through superstructure optimization or heuristic rules. To enable autonomous process design, it is necessary to formalize knowledge in a machine-readable format. This study proposes incorporating ontological representation of fundamental process knowledge to enhance general-purpose design procedures, while considering problem-specific variability. The framework leverages an ontology to express declarative knowledge and uses a hierarchical reinforcement learning agent to learn procedural knowledge, leading to more efficient and high-quality solutions. The case study on intensified steam methane reforming process demonstrates the benefits of automating domain knowledge in reducing search space and improving computational efficiency and solution quality, highlighting its potential in autonomous process design approaches.
COMPUTERS & CHEMICAL ENGINEERING
(2023)
Article
Chemistry, Multidisciplinary
Xiaochun Sun, Chenmou Wu, Shuqun Yang
Summary: With the proliferation of Knowledge Graphs (KGs), knowledge graph completion (KGC) has attracted much attention. Previous methods focused on extracting shallow structural information from KGs or combining with external knowledge. To address the limitations, a novel Scalable Formal Concept-driven Architecture (SFCA) was proposed to encode factual triples into formal concepts, providing valuable information for KGC. Comprehensive experiments on public datasets and industry dataset demonstrated the effectiveness and scalability of SFCA, offering new ideas for the promotion and application of knowledge graphs in AI downstream tasks.
APPLIED SCIENCES-BASEL
(2023)
Article
Automation & Control Systems
Agustin Borrego, Daniel Ayala, Inma Hernandez, Carlos R. Rivero, David Ruiz
Summary: This paper proposes an approach to complete Knowledge Graphs by evaluating candidate triples using neighborhood-based features, which outperforms other state-of-the-art methods in identifying correct triples and achieving higher average F1 scores in all tested datasets.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2021)
Article
Computer Science, Software Engineering
Mingwei Liu, Chengyuan Zhao, Xin Peng, Simin Yu, Haofen Wang, Chaofeng Sha
Summary: AI applications often use ML/DL models to implement specific tasks, but it is difficult for developers to find suitable ML/DL libraries due to the rapid development of AI and lack of detailed information about the libraries. This paper proposes a task-oriented recommendation approach called MLTaskKG, which utilizes a knowledge graph to recommend ML/DL libraries based on developers' requirements and characteristics of AI tasks and models. Evaluation results show the high quality and effectiveness of MLTaskKG in helping developers find suitable ML/DL libraries.
IEEE TRANSACTIONS ON SOFTWARE ENGINEERING
(2023)
Article
Genetics & Heredity
Bijun Zhang, Ting Fan
Summary: This study summarizes the current literature on the application of deep learning in genetics and analyzes the current research characteristics and future trajectories in this field. The study found that the number of publications on deep learning applications in genomics is rapidly increasing, with the United States having the highest research output and collaborating closely with China and Germany. Keyword evolution suggests that the current research frontiers in this field include prediction, sequence, mutation, and cancer.
FRONTIERS IN GENETICS
(2022)
Article
Engineering, Industrial
Ang Liu, Qiuyu Yu, Boming Xia, Qinghua Lu
Summary: This study discusses the dilemma between the effectiveness of machine learning and data privacy protection in the design of smart products, introducing federated learning as an emerging solution. An experiment validates the effectiveness of federated learning in predicting electricity consumption.
CIRP ANNALS-MANUFACTURING TECHNOLOGY
(2021)
Article
Computer Science, Artificial Intelligence
Xiang Lan, Yahong Hu, Youbai Xie, Xianghui Meng, Yilun Zhang, Qiangang Pan, Yishen Ding
Summary: Based on the basic law of Design Science, new product design relies on existing design knowledge. To improve design quality and reduce design time, knowledge integration is used in product function design by effectively utilizing existing knowledge. The complexity of product design and the abundance of design knowledge make it increasingly difficult for traditional traversal-based algorithms to complete knowledge integration within an acceptable time frame.
ADVANCED ENGINEERING INFORMATICS
(2023)
Article
Computer Science, Artificial Intelligence
Jinpu Zhang, Guozhong Cao, Qingjin Peng, Runhua Tan, Wei Liu, Huangao Zhang
Summary: This research focuses on building a case base of transformation parameters and principles to guide the design of transformable products. A systematic design process is proposed to apply the design knowledge effectively. The method is validated in the design of a self-propelled boom sprayer.
ADVANCED ENGINEERING INFORMATICS
(2022)
Article
Biology
Chaimae Asaad, Mounir Ghogho
Summary: This paper proposes an asthma knowledge graph (KG) built using a hybrid methodology, focusing on environmental interactions. By utilizing public sources and natural language processing, a genetic and pharmacogenetic asthma knowledge graph and an Asthma-Environment interaction catalog were constructed. The results were integrated into the knowledge graph.
COMPUTERS IN BIOLOGY AND MEDICINE
(2022)
Article
Computer Science, Hardware & Architecture
Kourosh Hakhamaneshi, Marcel Nassar, Mariano Phielipp, Pieter Abbeel, Vladimir Stojanovic
Summary: In this article, a supervised pretraining approach is proposed to learn circuit representations for predicting circuit performance, enabling automated design. By training a neural network to predict the output dc voltages of circuit instances, generalizable knowledge about the role of each circuit element and their interactions can be obtained. Graph neural networks are used to learn node embeddings by representing circuits as graphs, allowing adaptation to new topologies or prediction tasks.
IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Jia Jia, Yingzhong Zhang, Mohamed Saad
Summary: This paper proposes a scientific method to capture and reuse tacit design knowledge, using design knowledge graphs to represent the observable design result facts of products and employing relational learning to search for new design solution requirements. The research demonstrates that this method is highly feasible, effective, and flexible.
ADVANCED ENGINEERING INFORMATICS
(2022)
Article
Engineering, Multidisciplinary
Yuexin Huang, Suihuai Yu, Jianjie Chu, Zhaojing Su, Yangfan Cong, Hanyu Wang, Hao Fan
Summary: This study proposes a novel method that combines deep learning with knowledge graph to acquire valuable design knowledge in conceptual product design. The method utilizes a knowledge extraction model to extract design-related entities and relations from fragmentary data and constructs a knowledge graph to support design knowledge acquisition. Experimental comparison and case study both verify the effectiveness and feasibility of the proposed method, showcasing its potential in providing designers with interconnected and visualized design knowledge.
CMES-COMPUTER MODELING IN ENGINEERING & SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Wenqiang Liu, Hongyun Cai, Xu Cheng, Sifa Xie, Yipeng Yu, Dukehyzhang
Summary: The goal of representation learning of knowledge graph is to encode entities and relations into a low-dimensional embedding space. Existing methods have limitations in expressing high-order structural relationships between entities and utilizing attribute triples. To overcome these limitations, this paper proposes a novel method named KANE, which captures high-order structural and attribute information of knowledge graphs using graph convolutional networks. Experimental results show that KANE outperforms other methods.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Information Systems
Pouya Ghiasnezhad Omran, Kerry Taylor, Sergio Rodriguez Mendez, Armin Haller
Summary: This article introduces a novel algorithm, OPRL, for learning Open Path (OP) rules that can generate relevant queries for Knowledge Graph completion, even when there is no closed rule to answer the query. This demonstrates the first solution for active knowledge graph completion.
INFORMATION SCIENCES
(2022)
Article
Engineering, Industrial
Qinglin Qi, Fei Tao, Tianliang Hu, Nabil Anwer, Ang Liu, Yongli Wei, Lihui Wang, A. Y. C. Nee
Summary: Digital twin is revolutionizing industry by mirroring almost every facet of a product, process or service in the digital space. However, realizing their full potential is a complex process as researchers need to model different parts of objects or systems, collect and merge varied types of data, and struggle with determining the technologies and tools to be used. The 5-dimension digital twin model provides guidance for understanding and implementing digital twin technologies.
JOURNAL OF MANUFACTURING SYSTEMS
(2021)
Article
Engineering, Industrial
Xingzhi Wang, Yuchen Wang, Fei Tao, Ang Liu
Summary: Big data plays a crucial role in smart customisation, and the integration of physical and virtual entities through digital twin technology enhances responsiveness and predictability in the customisation process.
JOURNAL OF MANUFACTURING SYSTEMS
(2021)
Editorial Material
Engineering, Industrial
Fei Tao, Nabil Anwer, Ang Liu, Lihui Wang, Andrew Y. C. Nee, Liming Li, Meng Zhang
JOURNAL OF MANUFACTURING SYSTEMS
(2021)
Article
Engineering, Industrial
Ang Liu, Qiuyu Yu, Boming Xia, Qinghua Lu
Summary: This study discusses the dilemma between the effectiveness of machine learning and data privacy protection in the design of smart products, introducing federated learning as an emerging solution. An experiment validates the effectiveness of federated learning in predicting electricity consumption.
CIRP ANNALS-MANUFACTURING TECHNOLOGY
(2021)
Article
Automation & Control Systems
Lei Wang, Zhengchao Liu, Ang Liu, Fei Tao
Summary: Artificial intelligence technology is gaining increasing attention in the field of smart manufacturing, particularly in product lifecycle management. This paper examines the theories, algorithms, and technologies of AI in PLM and presents a structured roadmap for future research and application. The opportunities and challenges of applying AI to PLM are also discussed.
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
(2021)
Article
Business
Zhinan Zhang, Jun Liu, Richard David Evans, Ang Liu
Summary: The global manufacturing industry is experiencing growth and innovation due to advancements in technology. Effective communication is key to exchanging knowledge among stakeholders with different backgrounds, especially in product design. By proposing a method to enhance understanding of target objects and a knowledge representation framework, communication efficiency can be greatly improved in design collaboration.
IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT
(2021)
Article
Automation & Control Systems
Shikang Chen, Xinjiang Cai, Xingzhi Wang, Ang Liu, Qinghua Lu, Xiwei Xu, Fei Tao
Summary: Blockchain, as a new generation of information technology, possesses unique traits such as data traceability, security, decentralized consensus, and execution. This technology is crucial for smart manufacturing, especially in product lifecycle management. The paper outlines the key features and history of blockchain, examines challenges in data management and security in PLM, and proposes using blockchain to address these issues.
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
(2022)
Article
Computer Science, Interdisciplinary Applications
Haiyang Zhao, Liangchi Zhang, Zhonghuai Wu, Ang Liu
Summary: This paper presents a new discrete element model for rock-like materials, which is superior to most models in algorithm simplicity and stress analysis. The model is more effective in predicting crack behavior of rock-like materials.
COMPUTERS & STRUCTURES
(2022)
Proceedings Paper
Computer Science, Information Systems
Dawen Zhang, Xiwei Xu, Liming Zhu, Hye-Young Paik
Summary: Blockchain technology has been applied to various industries' supply chains, providing a level playing field for small-to-medium size businesses to participate while improving supply chain integrity. However, significant uncertainties exist in supply chains at both physical and digital levels, necessitating flexible mechanisms for run-time process adaptation. The proposed framework in this paper allows for dynamic attachment and detachment of process fragments defined in smart contracts to enable process adaptation.
2021 IEEE INTERNATIONAL CONFERENCE ON BLOCKCHAIN AND CRYPTOCURRENCY (ICBC)
(2021)
Article
Engineering, Manufacturing
Ang Liu, Dawen Zhang, Xingzhi Wang, Xiwei Xu
Summary: This paper proposes a blockchain-based collaborative customization framework to enhance collaborations among stakeholders in product customization. The framework improves transparency, flexibility, efficiency, and traceability, and a practical example demonstrates its application in elevator customization.
MANUFACTURING LETTERS
(2021)
Article
Social Sciences, Interdisciplinary
Khaled Medini, Sophie Peillon, Martha Orellano, Stefan Wiesner, Ang Liu
Summary: The evolution toward more customer-centric operations in manufacturing and service industries has led to the emergence of Product-Service Systems (PSS) as a novel way of value creation and delivery. Close collaboration among various actors in a value network is crucial for creating win-win gains. However, the lack of economic assessment models and methods specific to PSS poses a challenge.