Article
Computer Science, Artificial Intelligence
Ting-Hsiang Wang, Hsiu-Wei Yang, Chih-Ming Chen, Ming-Feng Tsai, Chuan-Ju Wang
Summary: This paper investigates item concept modeling and proposes a framework called ICN, which leverages the inferential property of concepts to address the issue of conceptual correlation sparsity. The framework consists of two stages: ICN construction and embedding learning, and it outperforms traditional methods in item classification and retrieval tasks.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2022)
Article
Neurosciences
Yi-Chieh Chiu, Tracy H. Wang, Diane M. Beck, Jarrod A. Lewis-Peacock, Lili Sahakyan
Summary: This study shows that context processing is crucial in the forgetting of individual items, with scene-related activity increasing and item-related information decreasing after a forget instruction. The separation between item information and context information predicts successful forgetting.
Article
Mathematics
K. Jeganathan, S. Selvakumar, S. Saravanan, N. Anbazhagan, S. Amutha, Woong Cho, Gyanendra Prasad Joshi, Joohan Ryoo
Summary: This paper discusses an integrated and interconnected stochastic queuing-inventory system that provides multi-type service to multi-class customers through a dedicated channel. The system aims to increase the occurrence of different types of customers by offering a multi-type service facility in one place. The results show that allowing customers to sell used items and purchase new items in the system assumption will increase the number of customers approaching the system.
Article
Automation & Control Systems
Ming Li, Yan Song
Summary: The proposed TF-like SNLF model achieves a good compromise between model performance and time efficiency by introducing LF matrix and fixed weighting matrix.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Jia Liu, Wei Huang, Tianrui Li, Shenggong Ji, Junbo Zhang
Summary: This paper proposes a multi-domain item-item recommendation method based on cross-domain knowledge graph embedding, which addresses the sparsity and cold start problems faced by traditional recommender systems by analyzing the association between items within the same domain and the interaction between items across diverse domains with the aid of a rich information knowledge graph.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Weiqiang Di, Zhihao Wu, Youfang Lin
Summary: Sequential recommendation is a research trend that uses user's recent behaviors for recommendation. This paper proposes an improved item-item product (IIP) mechanism that considers position and decay of historical item influence to enhance recommendation performance.
PEERJ COMPUTER SCIENCE
(2022)
Article
Multidisciplinary Sciences
Yeajin Ham, Suyeong Bae, Heerim Lee, Yaena Ha, Heesu Choi, Ji-Hyuk F. Park, Hae Yean F. Park, Ickpyo F. Hong
Summary: This study evaluated the item-level psychometrics of the AD-8 through confirmatory factor analysis and the Rasch rating scale model. The results showed that the AD-8 has a unidimensional measurement structure, no misfitting items, and no differential item functioning across sex and gender. Despite floor effects, the AD-8 demonstrated good reliability and can be used as a screening assessment tool in clinical settings.
Article
Computer Science, Information Systems
Hanrui Wu, Chung Wang Wong, Jia Zhang, Yuguang Yan, Dahai Yu, Jinyi Long, Michael K. Ng
Summary: Recommendation systems personalize service for users by suggesting items they may prefer. This article focuses on next-item recommendation systems under the cold-start situation, where users have no interaction with new items. It proposes a novel model called User-Item Matching and Auto-encoders (UIMA) that learns latent embeddings for users and items through user preferences and item attributes, and explores the relationship between them using a matching network.
IEEE TRANSACTIONS ON SERVICES COMPUTING
(2023)
Article
Computer Science, Information Systems
Haiyang Zhang, Ivan Ganchev, Nikola S. Nikolov, Zhanlin Ji, Mairtin O'Droma
Summary: Matrix Factorization is a successful Collaborative Filtering technique in recommender systems and incorporating item features in a single step process has been shown to significantly improve recommendation performance. The proposed model, FeatureMF, enriches item representation in MF by projecting item features into the same latent factor space with users and items, yielding the best recommendation performance across all contexts and effectively alleviating data sparsity and cold-start issues. The model is also found to scale well in terms of computational time with increasing dataset size.
Article
Computer Science, Information Systems
Hideo Hirose
Summary: This paper investigates the performance of item response theory based on distance criteria rather than likelihood criteria. A reconstructed item response matrix using maximum likelihood estimates is introduced, and the distance between the observed and estimated matrices is measured using the Frobenius matrix norm. The comparison of the distance between the observed and low-rank matrices helps evaluate the performance of the estimated item response matrix. The study finds that the predictive ability of item response theory is high enough when using test data, as the distance between the approximated low-rank matrix and observed item response matrix is approximately equal to or slightly less than the distance between the estimated item response matrix and observed item response matrix.
Article
Education & Educational Research
R. Westacott, K. Badger, D. Kluth, M. Gurnell, M. W. R. Reed, A. H. Sam
Summary: This study used computer software to generate multiple item variants and assessed them with final year medical students in the UK. The results showed that there were significant differences in item facility, which may be related to different clinical reasoning strategies and school-level factors.
BMC MEDICAL EDUCATION
(2023)
Review
Psychology
Gordon D. Logan, Gregory E. Cox
Summary: In response to Osth and Hurlstone's (2022) commentary, this article addresses four issues related to the context retrieval and updating (CRU) theory of serial order proposed by Logan (2021). The authors clarify the relationships between CRU, chains, and associations, highlighting that CRU relies on similarity rather than association for context retrieval. They also correct an error in Logan's (2021) explanation for the tendency to recall ACB instead of ACD in the ABCDEF sequence. Furthermore, the article discusses position-specific prior-list intrusions and provides modifications to CRU, introducing a position-coding model based on CRU representations. The authors suggest that position-specific between-group intrusions in structured lists cannot be accounted for by reasonable modifications to CRU, but these intrusions still support position coding on some trials without ruling out CRU-like item-based codes. The article emphasizes the importance of considering immediate performance and presents item-independent and item-dependent coding as alternative strategies for serial recall.
PSYCHOLOGICAL REVIEW
(2023)
Article
Engineering, Industrial
Karim Tamssaouet, Erna Engebrethsen, Stephane Dauzere-Peres
Summary: This paper addresses the tactical joint inventory and transportation planning problem for multiple items with deterministic and time-varying demand, considering different transportation modes and item fragmentation. The proposed problem tackles the conflict between potentially reducing the number of containers used and negatively impacting handling and shipping operations. Several Mixed Integer Linear Programming models are suggested and a relax-and-fix heuristic is proposed for solving the problem. Realistic instances are used for computational experiments to identify the most efficient model, study the impact of key parameters, and analyze the efficiency of the heuristic. Managerial insights are derived to identify contexts requiring joint optimization and the impact of item fragmentation constraints, and future research directions are proposed.
INTERNATIONAL JOURNAL OF PRODUCTION ECONOMICS
(2023)
Article
Multidisciplinary Sciences
Raul Costa Mastrascusa, Matheus Loli de Oliveira Fenili Antunes, Nathalia Saraiva de Albuquerque, Sara Luisa Virissimo, Marcela Foletto Moura, Bibiana Vieira Marques Motta, Wagner de Lara Machado, Carmen Moret-Tatay, Tatiana Quarti Irigaray
Summary: The main purpose of this study was to evaluate the factorial structure and reliability of the 44-item BFI as well as its shorter versions with 20 and 10 items. The study also aimed to provide normative data for interpreting scores from the short and ultrashort versions of the BFI for the Brazilian population. The results showed that the original 44-item model had poor adaptation, but the shorter versions with 20 and 10 items had good adaptation indexes and reliability.
SCIENTIFIC REPORTS
(2023)
Article
Psychology, Multidisciplinary
Qishan Chen, Haiyan Zheng, Honglan Fan, Lei Mo
Summary: The purpose of this study was to construct and validate a comparable item bank for assessing fourth-grade students' reading literacy. A total of 115 reading comprehension items were developed and administered to construct the item bank. The final item bank included 99 reading performance indicators and showed good psychometric characteristics, indicating its usefulness in assessing the reading literacy of fourth graders.
FRONTIERS IN PSYCHOLOGY
(2023)
