Article
Physics, Multidisciplinary
Shengteng Jiang, Yueling Liu, Yichi Zhang, Peng Luo, Kuo Cao, Jun Xiong, Haitao Zhao, Jibo Wei
Summary: Semantic communication is a promising technology for addressing the challenges of large bandwidth and power requirements caused by data explosion. By using a knowledge graph, it improves the accuracy of semantic representation, removes semantic ambiguity, and significantly enhances communication reliability in low signal-to-noise conditions.
Article
Computer Science, Artificial Intelligence
Raabia Mumtaz, Muhammad Abdul Qadir
Summary: This paper introduces a system CustRE for identifying and classifying family relations from English text. By using rules, regular expressions, and co-reference rules, it successfully extracts explicit and implicit family relations mentioned in the text.
KNOWLEDGE AND INFORMATION SYSTEMS
(2022)
Article
Health Care Sciences & Services
Kai He, Lixia Yao, JiaWei Zhang, Yufei Li, Chen Li
Summary: Researchers utilized online obituary data to construct genealogical knowledge graphs, successfully extracting and assembling family relationship data through a multitask neural network model, providing more comprehensive and accurate support for biomedical research.
JOURNAL OF MEDICAL INTERNET RESEARCH
(2021)
Article
Computer Science, Artificial Intelligence
Bogdan Walek, Petr Fajmon
Summary: This article proposes a hybrid recommender system that combines collaborative filtering, content-based approaches, and a fuzzy expert system. By analyzing user preferences and activity, and using the fuzzy expert system to create a recommended product list, this system achieves promising results based on standard metrics, even outperforming traditional approaches.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Review
Construction & Building Technology
Hui Deng, Yiwen Xu, Yichuan Deng, Jiarui Lin
Summary: This paper reviews and summarizes the application of ICTs in knowledge management in the architecture, engineering and construction industry. It reveals the development line of key technologies and discusses the advantages and disadvantages of different ICTs in knowledge management processes. It also identifies the imbalance in industry and academia, as well as cognitive barriers and lack of evaluation standards. Suggestions for future development are proposed.
AUTOMATION IN CONSTRUCTION
(2022)
Article
Environmental Sciences
Shahryar Sarabi, Qi Han, Bauke de Vries, A. Georges L. Romme, Dora Almassy
Summary: Deriving knowledge from past experiences is crucial for the successful adoption of NBS. However, extracting knowledge from the vast amount of information provided by repositories is challenging. This paper introduces an expert system called NBS-CBS, which combines a black-box artificial neural networks model with a white-box case-based reasoning model to provide intelligent, adaptive, and explainable recommendations and information for NBS planning and design.
JOURNAL OF ENVIRONMENTAL MANAGEMENT
(2022)
Review
Computer Science, Artificial Intelligence
Prantika Chakraborty, Debarshi Kumar Sanyal
Summary: Personal knowledge graphs (PKGs) are knowledge graphs that store detailed information relevant to a user but not generally useful to the rest of humanity. They offer more useful insights than general knowledge for personalized tasks. Despite their wide applicability, there is limited research in this area. This article provides an extensive review of PKGs, categorizing them based on relevant domains and discussing limitations and future directions.
WILEY INTERDISCIPLINARY REVIEWS-DATA MINING AND KNOWLEDGE DISCOVERY
(2023)
Article
Chemistry, Multidisciplinary
Wei Chen, Yihao Zhang, Yantuan Xian, Yonghua Wen
Summary: This paper introduces a recommendation model called PRHN to address the information overload problem in literature searches for academic articles. By analyzing the connections between literature, domain knowledge, and hotspot information, PRHN can better locate relevant documents for domain literature. Compared to previous models, PRHN demonstrates improved HR and NDCG on public available datasets.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Artificial Intelligence
Wenhui Liao, Qian Zhang, Bo Yuan, Guangquan Zhang, Jie Lu
Summary: This article proposes a novel multidomain recommender system called HMRec to address the challenges of exploiting valuable information from multiple source domains and ensuring positive transfer from heterogeneous data. By extracting domain-shared and domain-specific features, HMRec enables knowledge transfer between multiple heterogeneous source and target domains. Extensive experiments demonstrate that HMRec effectively increases rating prediction accuracy in the target domain and outperforms six state-of-the-art baselines.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Sirichanya Chanmee, Kraisak Kesorn
Summary: This study introduces a new approach called the Semantic Decision Tree (SDT) to effectively address the multi-value bias selection issue and improve the generation of decision tree nodes. Evaluation results on multiple datasets show that SDT outperforms traditional algorithms in terms of accuracy and aligns more naturally with human decision-making logic.
ADVANCED ENGINEERING INFORMATICS
(2023)
Article
Computer Science, Artificial Intelligence
Thien Khai Tran, Chien D C Ta, Tuoi Thi Phan
Summary: Semantic relations have been applied in various research fields to eliminate conceptual and terminological confusion. This research describes the detection of semantic relations using WordNet and entities from a knowledge graph, achieved through natural language processing and deep learning. The generated knowledge graph shows excellent performance, as demonstrated in the evaluation using five categories from the ACM Digital Library.
JOURNAL OF INTELLIGENT & FUZZY SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Hong Yin, Jiang Zhong, Chen Wang, Rongzhen Li, Xue Li
Summary: Knowledge graph completion aims to infer missing links between entities based on observed ones. Current methods focus on KG embedding models but suffer from limitations in considering global semantic and interactions between neighbors. To address this, we propose GS-InGAT, a KGC method that combines semantic graph modeling with an Interaction Graph ATtention network (InGAT) to capture both interaction and local information. Experimental results demonstrate the effectiveness of considering global semantic and interactions, showcasing the comparable performance of GS-InGAT on benchmark datasets.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Nagendra Kumar, Eshwanth Baskaran, Anand Konjengbam, Manish Singh
Summary: This study presents a novel hashtag recommendation method that addresses data sparsity by utilizing external knowledge sources, incorporating lexical, topical, semantic, and user influence features, aggregating various recommendation methods using learning-to-rank, and outperforming current state-of-the-art methods according to experimental results.
KNOWLEDGE AND INFORMATION SYSTEMS
(2021)
Article
Computer Science, Information Systems
Daksh Dave, Aditya Sharma, Shafi'i Muhammad Abdulhamid, Adeel Ahmed, Adnan Akhunzada, Rashid Amin
Summary: Due to the rapid growth of mobile applications and users' difficulty in finding suitable apps, app recommender systems have emerged as a helpful tool. However, these systems require access to user data, posing a serious security violation. To address this issue, we developed SAppKG, a user privacy-preserving knowledge graph architecture for mobile app recommendation. Testing it on real-world data from the Google Play store, we found that SAppKG improved results on various metrics compared to baseline models.
Article
Computer Science, Artificial Intelligence
Junlin Zhu, Jiaye Wu, Xudong Luo, Jie Liu
Summary: In this paper, an intelligent system for legal information retrieval on the WeChat platform is developed to provide efficient and convenient legal knowledge services during the COVID-19 pandemic. The system is trained using typical cases published online by the Supreme People's Procuratorate of China. It utilizes convolutional neural network and semantic matching mechanism to capture inter-sentence relationship information and make predictions. Additionally, an auxiliary learning process is introduced to help the network better distinguish the relation between two sentences. The system uses the trained model to identify user input and responds with a reference case and legal gist applicable to the query case.
ARTIFICIAL INTELLIGENCE AND LAW
(2023)
