Article
Computer Science, Theory & Methods
Haosong Li, Phillip C-Y Sheu
Summary: This paper proposes a scalable association rule learning algorithm for efficiently learning gene association rules from large-scale microarray datasets. The algorithm ranks the rules based on their importance and outperforms the traditional Apriori algorithm in terms of performance.
JOURNAL OF BIG DATA
(2022)
Article
Computer Science, Artificial Intelligence
Gang Wang, Han-Ru Wang, Ying Yang, Dong-Ling Xu, Jian-Bo Yang, Feng Yue
Summary: Group Recommendation Systems (GRS) is an emerging area in research and practice, and this paper proposes a novel approach GPRAH_ER for group article recommendation using Probabilistic Matrix Factorization and ER rule to improve individual prediction and group aggregation processes, which outperforms benchmark methods in experiments on a real dataset.
APPLIED SOFT COMPUTING
(2021)
Article
Computer Science, Theory & Methods
Erna Hikmawati, Nur Ulfa Maulidevi, Kridanto Surendro
Summary: Association rule mining is a widely used technique in data mining to identify interesting relationships between sets of items in a dataset, with the key to success being the correct setting of thresholds. The proposed adaptive support method can generate rules according to user needs based on dataset characteristics.
JOURNAL OF BIG DATA
(2021)
Article
Biochemical Research Methods
Maidi Liu, Yanqing Ye, Jiang Jiang, Kewei Yang
Summary: This study proposed a microbial association network inference method 'MANIEA' based on the improved Eclat algorithm for mining positive and negative microbial association rules, and a new method for transforming association rules into microbial association networks. Experimental results demonstrated that 'MANIEA' has advantages in correlation forms, computation efficiency, adjustability, and network characteristics compared to currently popular network inference methods.
Article
Computer Science, Artificial Intelligence
Maidi Liu, Zhiwei Yang, Yong Guo, Jiang Jiang, Kewei Yang
Summary: Association rule mining (ARM) is an important research topic in data mining and knowledge discovery. This paper proposes a nonlinear ARM method called MICAR based on the maximal information coefficient (MIC), which can effectively extract high-quality positive and negative association rules, especially nonlinear association rules.
KNOWLEDGE AND INFORMATION SYSTEMS
(2022)
Article
Engineering, Industrial
He Lan, Xiaoxue Ma, Laihao Ma, Weiliang Qiao
Summary: Total loss of a ship is the most serious consequence of maritime accidents, causing massive property losses, human casualties, and environmental pollution. This study investigates significant patterns in total loss accidents using association rule technique and finds that ship age and accident type are key indicators.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2023)
Article
Computer Science, Cybernetics
Lina Wang, Xuyun Zhang, Tian Wang, Shaohua Wan, Gautam Srivastava, Shaoning Pang, Lianyong Qi
Summary: The article discusses the advantages and limitations of neighborhood-based collaborative filtering recommendation methods and proposes a diversified and scalable recommendation method that combines locality-sensitive hashing and cover tree to optimize the recommended list. The effectiveness and feasibility of the proposed method are demonstrated through experiments on the MovieLens dataset.
IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS
(2021)
Article
Computer Science, Cybernetics
Taushif Anwar, V Uma, Gautam Srivastava
Summary: This article addresses two research topics in recommender systems (RSs), namely cross-domain RSs (CDRSs) and context-aware RSs (CARSs). CDRSs aim to improve recommendation quality in a target domain by utilizing source domain information, while also combating the spread of fake information. CARSs utilize contextual information to recommend items that align with user interests as they change with context.
IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS
(2023)
Article
Chemistry, Multidisciplinary
Gui Li, Fan Liu, Cheng Wu, Yuan Yao, Guangxin Wu, Zhu Wang, Yanchun Zhang
Summary: This paper proposes a classification framework based on multiple weighted class association rules (C-MWCAR) to improve classification performance. The framework includes a CAR mining algorithm, a CAR selection algorithm, and a weighted CARs-based classifier. Experimental results demonstrate that C-MWCAR outperforms four baseline methods.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Software Engineering
Meryem Uzun-Per, Ahmet Volkan Gurel, Ali Burak Can, Mehmet S. Aktas
Summary: In this paper, a recommendation system framework based on Lambda architecture is proposed for recommending ancillary services for airline companies. The proposed method utilizes association rule and sequential pattern mining algorithms on big data processing platforms. Experimental results demonstrate the usefulness and negligible processing overheads of the proposed method.
CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE
(2022)
Article
Computer Science, Theory & Methods
Haosong Li, Phillip C-Y Sheu
Summary: This paper presents a heuristic approach based on divide-and-conquer to address the scalability issues in association rule learning, showing significant speedup and approximate results close to accurate results compared to existing algorithms.
JOURNAL OF BIG DATA
(2021)
Article
Computer Science, Information Systems
Diana Sola, Han van der Aa, Christian Meilicke, Heiner Stuckenschmidt
Summary: Business process modeling is crucial in organizations, but creating consistent and complete process models is challenging. This paper proposes a rule-based and semantic-aware recommendation approach to improve the quality of activity label recommendations.
INFORMATION SYSTEMS
(2022)
Article
Construction & Building Technology
Yan Li, Zhengbo Zou, Jiupeng Zhang, Yinzhang He
Summary: In order to study the relationship between distresses on airfield asphalt pavements, 500 units of asphalt runway data from five airports were collected and surveyed using a scientific pavement zoning method. An improved Apriori algorithm was used to explore the evolution mechanism between typical distresses and establish a pavement evolution model. The results showed three stages of distress evolution and that each distress could evolve from the previous one. Delaying and stopping the process of distress evolution and accurately identifying the key points of evolution are crucial for extending the service life of asphalt pavements.
CONSTRUCTION AND BUILDING MATERIALS
(2023)
Article
Computer Science, Hardware & Architecture
Chunjing Xiao, Shenkai Lv, Wei Fan, W. H. Ip
Summary: In this paper, a dynamic graph evolution model is proposed to accurately predict user preference. The model captures temporal-order item associations and user relationships, and updates embeddings and graphs through a dynamic evolution mechanism based on two GCNs. Experimental results show that the proposed method outperforms state-of-the-art methods on real-world datasets.
JOURNAL OF SUPERCOMPUTING
(2023)
Article
Business
Onur Dogan
Summary: E-commerce is rapidly growing, making it crucial to understand complex transactional data in order to provide practical product recommendations. Traditional methods focus on recommending frequent items, but fail to consider profitability. This study introduces a novel method called P-FARM, which mines association rules based on profitability and frequent item sets. The results show that P-FARM is a powerful tool for improving e-commerce sales and maximizing profit.
JOURNAL OF THEORETICAL AND APPLIED ELECTRONIC COMMERCE RESEARCH
(2023)