Article
Computer Science, Interdisciplinary Applications
Yassine Afoudi, Mohamed Lazaar, Mohammed Al Achhab
Summary: Recommendation systems are tools that provide information based on user preferences and behavior, utilizing methods like Collaborative Filtering, Content Based Approach, and neural network techniques. Research shows that a hybrid recommender framework method improves accuracy and efficiency compared to traditional Collaborative Filtering methods.
SIMULATION MODELLING PRACTICE AND THEORY
(2021)
Article
Computer Science, Artificial Intelligence
Yuyuan Li, Chaochao Chen, Xiaolin Zheng, Yizhao Zhang, Biao Gong, Jun Wang, Linxun Chen
Summary: Recent regulations on the Right to be Forgotten have significantly impacted recommender systems by allowing users to withdraw their private data. This paper proposes a highly efficient recommendation unlearning method, SCIF, that avoids retraining and preserves collaboration between users and items. The method is evaluated using a Membership Inference Oracle to assess unlearning completeness. Experimental results demonstrate that the proposed method improves efficiency and outperforms existing methods in comprehensive recommendation metrics.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Celestine Iwendi, Ebuka Ibeke, Harshini Eggoni, Sreerajavenkatareddy Velagala, Gautam Srivastava
Summary: The creation of digital marketing has enabled companies to adopt personalized item recommendations, improving their competitive advantage. This paper proposes a machine learning model system that introduces a new rating system and combines it with existing review systems, resulting in improved recommendation accuracy and effectiveness.
INTERNATIONAL JOURNAL OF INFORMATION TECHNOLOGY & DECISION MAKING
(2022)
Article
Computer Science, Artificial Intelligence
Yuefang Gao, Zhen-Wei Huang, Zi-Yuan Huang, Ling Huang, Yingjie Kuang, Xiaojun Yang
Summary: Recently, neighborhood-based collaborative filtering has been used more and more in personalized recommender systems. However, the traditional approach of selecting a fixed number of nearest users/items as neighbors has limitations. To address this issue, a new recommender system called Multi-scale Broad Collaborative Filtering (MBCF) is proposed, which captures rich information from different numbers of nearest users/items. Instead of using deep neural networks (DNNs), the Broad Learning System (BLS) is adopted to learn the complex nonlinear relationships between users and items, achieving satisfactory recommendation performance while avoiding overfitting. Extensive experiments on eight benchmark datasets demonstrate the effectiveness of the proposed MBCF algorithm.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Chemistry, Multidisciplinary
Raushan Kumar Singh, Pradeep Kumar Singh, Juginder Pal Singh, Akhilesh Kumar Singh, Seshathiri Dhanasekaran
Summary: Collaborative filtering is the most popular method for addressing information overload in E-Commerce. Traditional collaborative filtering predicts the target item based on ratings from similar users. However, similarity calculation in sparse datasets may lead to decreased performance. This study proposes a new approach that considers item features to improve accuracy, using ratings from individuals with the most similar features instead of relying on the wisdom of the crowd.
APPLIED SCIENCES-BASEL
(2022)
Article
Computer Science, Information Systems
Shitao Xiao, Yingxia Shao, Yawen Li, Hongzhi Yin, Yanyan Shen, Bin Cui
Summary: The paper introduces a novel collaborative filtering framework, LECF, which models interactions between users and items as edges and captures complex relationships. LECF predicts the existence probability of edges based on weighted similarities in a line graph and utilizes an efficient propagation algorithm for training and inference. Experimental results demonstrate that LECF outperforms state-of-the-art methods on four public datasets.
SCIENCE CHINA-INFORMATION SCIENCES
(2022)
Article
Mathematics
Zhiqiang Pan, Honghui Chen
Summary: This research introduces the Collaborative Knowledge-Enhanced Recommendation (CKER) method, which utilizes a collaborative graph convolution network (CGCN) to learn user and item representations and incorporates self-supervised learning to maximize mutual information between user preferences. Experimental results demonstrate that CKER outperforms state-of-the-art baselines in the field of knowledge-enhanced recommendation.
Article
Computer Science, Artificial Intelligence
Bingbing Dong, Yi Zhu, Lei Li, Xindong Wu
Summary: Personalized recommendation systems have been a focus of attention in recent decades for recommending products and services to users. The proposed Item-Agrec model utilizes a semi-autoencoder to co-embed the attributes and graph features of items, improving recommendation accuracy.
Article
Computer Science, Information Systems
Urvish Thakker, Ruhi Patel, Manan Shah
Summary: Collaborative Filtering (CF) is a widely used technology in recommender systems, which effectively utilizes information from applications to find similarities and has been applied in various industries. This paper discusses the algorithm and applications of CF in Movie Recommendation System (MRS), as well as challenges and future developments in the field.
MULTIMEDIA TOOLS AND APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Baboucarr Drammeh, Hui Li
Summary: In the past decade, there has been significant growth in online transactions. As a result, many professionals and researchers have turned to deep learning models to design and develop recommender systems for online personal services. However, existing approaches often fail to accurately represent the correlation between users and items. Therefore, this article proposes a deep collaborative recommendation system based on a convolutional neural network, which incorporates an outer product matrix and a hybrid feature selection module to capture both local and global higher-order interactions.
PEERJ COMPUTER SCIENCE
(2023)
Article
Computer Science, Artificial Intelligence
Yihao Zhang, Xiaoyang Liu
Summary: Deep learning has been dominating the recommender system field, but there are shortcomings in existing methods such as handling of user-item interactions and long-term sequential dependencies. To address these issues, a novel NCRAE algorithm based on memory networks is proposed, which can better learn attention embeddings from user-item interactions and improve recommendation performance.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Chemistry, Multidisciplinary
Ninghua Sun, Tao Chen, Wenshan Guo, Longya Ran
Summary: This study proposed a method to enhance CF-based recommender system by utilizing negative item information and original embedding features to learn latent features of positive items. Experimental results demonstrated significant performance improvement compared with state-of-the-art baseline algorithms.
APPLIED SCIENCES-BASEL
(2021)
Article
Engineering, Multidisciplinary
Shenghao Liu, Bang Wang, Xianjun Deng, Laurence T. Yang
Summary: This paper proposes a novel recommendation algorithm that combines self-attentive graph convolution network, latent group mining, and collaborative filtering. Experimental results show that the algorithm outperforms the state-of-the-art algorithms on real-world datasets.
IEEE TRANSACTIONS ON NETWORK SCIENCE AND ENGINEERING
(2022)
Article
Computer Science, Information Systems
Le Nguyen Hoai Nam
Summary: This paper focuses on the rating prediction phase in memory-based collaborative filtering and improves the prediction accuracy by optimizing an objective function. Experimental results demonstrate that the proposed method outperforms others, especially when the number of selected neighbors is small to medium.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Artificial Intelligence
Qingxian Wang, Suqiang Wu, Yanan Bai, Quanliang Liu, Xiaoyu Shi
Summary: The emerging topic of Graph Neural Networks (GNN) has achieved state-of-the-art performance in recommendation problems due to its strong ability in node representation. We propose BIG-SAGE@, a neighbor importance-aware GNN, for item recommendation and rating prediction. Through rating confidence-based neighborhood sampling and an attention network, BIG-SAGE@ outperforms SOTA methods in rating prediction and TopN ranking tasks.
Article
Chemistry, Applied
Peng Zhai, Chen Liu, Gang Feng, Yuanhao Cao, Lei Xiang, Keshi Zhou, Ping Guo, Jianqing Li, Wenxiao Jiang
Summary: In this study, a novel fluorescent microsphere-based flow-through immunoaffinity chromatographic assay was developed to detect ultratrace microcystin-LR in water and aquatic products. The assay allowed for quantitation of microcystin-LR in less than 30 minutes using fluorescent microspheres encapsulating aggregation-induced emission luminogens. Colorimetric images were captured and analyzed, providing a limit of detection of 0.217 pg/mL and a limit of quantitation of 0.362 pg/mL in water and aquatic muscle samples. The developed immunoassays showed high average recovery rates and low relative standard deviations, making them a practical on-site screening method for microcystin-LR at picogram levels in water and aquatic samples.
Article
Automation & Control Systems
Tingting Wang, Kai Fang, Wei Wei, Jinyu Tian, Yuanyuan Pan, Jianqing Li
Summary: This article proposes a microcontroller unit (MCU) chip temperature fingerprint informed machine learning method, called MTID, for Industrial Internet of Things (IIoT) intrusion detection. The method records the MCU temperature sequence of the node, analyzes the relationship between the temperature sequence and the computational complexity of the node, calculates temperature residuals, and constructs a temperature residuals dataset. A self-encoder-based intrusion detection model is then constructed to identify the security status of the nodes. Experimental analysis using Raspberry Pi 4B shows that the accuracy of MTID for intrusion detection reaches 89%.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Computer Science, Information Systems
Ying Sun, Xiaochen Yuan, Xingrun Wang, Jianqing Li
Summary: The paper introduces a grayscale-invariant multiple watermarking mechanism for color images, including multi-level watermarking mechanism and multi-region watermarking mechanism, to increase embedding capacity and reduce impact on the whole image.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Optics
Lin Gu, Zhen Liu, Yiqing Shu, Zhengwei Cui, Kefan Zhou, Jianqing Li, Aiping Luo, Weicheng Chen
Summary: In this study, we experimentally demonstrate the optical manipulation of h-shaped pulse generation in passively mode-locked fiber lasers by constructing composite filtering functions. Different types of h-shaped pulses can be dynamically manipulated by manipulating the superposition or multiplication between different independent filtering effects in the laser cavity.
OPTICS AND LASER TECHNOLOGY
(2023)
Article
Computer Science, Artificial Intelligence
Tingting Wang, Bingxian Lu, Wei Wang, Wei Wei, Xiaochen Yuan, Jianqing Li
Summary: This paper presents a game theory-based distributed edge computing server task scheduling model that balances link quality and computing resource requirements, provides differentiated services for different priority users, accurately predicts link quality using a time series prediction algorithm, and achieves Nash equilibrium quickly through an acceleration scheme.
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE
(2023)
Article
Engineering, Environmental
Xiangping Lin, Zhongjun Li, Shuang Du, Qun Wang, Yucheng Guan, Guopan Cheng, Huijie Hong, Jianqing Li, Xiaojia Chen, Tongkai Chen
Summary: Breast cancer is a leading cause of cancer-related mortality in women and has a high rate of lung metastasis. To combat breast cancer and lung metastasis, researchers have developed a nanocomposite (Nb2C-BBR) based on functionalized niobium carbide nanosheets and berberine. The nanosheets have excellent photothermal conversion properties and drug delivery capacity, making them ideal for photothermal therapy. The nanocomposite showed favorable biocompatibility, reduced the required dose of nanosheets, and effectively destroyed cancer cells, inhibited metastasis, and induced minimal tissue damage, demonstrating its potential as a treatment approach for metastatic tumors.
CHEMICAL ENGINEERING JOURNAL
(2023)
Article
Computer Science, Artificial Intelligence
Xingrun Wang, Xiaochen Yuan, Mianjie Li, Ying Sun, Jinyu Tian, Hongfei Guo, Jianqing Li
Summary: This paper proposes a parallel multiple watermarking method using adaptive inter-block correlation. By considering the image texture characteristics and using a texture complexity method, suitable adjacent blocks are selected for watermark embedding, ensuring good imperceptibility and robustness. The proposed scheme optimizes the watermark embedding algorithm to achieve parallel embedding of multiple watermarks, reducing time cost.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Optics
Yingfang Zhang, Zhihao Lan, Liyazhou Hu, Yiqing Shu, Xun Yuan, Penglai Guo, Xiaoling Peng, Weicheng Chen, Jianqing LI
Summary: In this work, the existence of chiral topological electromagnetic edge states in Penrose-tiled photonic quasicrystals made of magneto-optical materials is demonstrated, without relying on the concept of bulk Bloch bands in momentum space. Despite the absence of bulk Bloch bands, some bandgaps in these photonic quasicrystals still could host unidirectional topological electromagnetic edge states immune to backscattering. Employing a real-space topological invariant based on the Bott index, it is shown that the bandgaps hosting these chiral topological edge states possess a nontrivial Bott index, depending on the direction of the external magnetic field. This work opens new possibilities for the study of topological states in photonic quasicrystals.
Article
Engineering, Environmental
A. Yuan, B. Wang, J. Li, Joseph H. W. Lee
Summary: Harmful Algal Blooms (HAB) have negative impacts on ecosystem functions and pose challenges to environmental and fisheries management. The development of real-time monitoring systems for algae populations and species is crucial for HAB management. This study proposes an on-site AI algae monitoring system with the AMDNN model embedded in an edge AI chip for real-time algae species classification and HAB prediction.
Article
Optics
Lin Gu, Zhen Liu, Yiqing Shu, Jianqing Li, Weicheng Chen
Summary: In this study, we experimentally synthesized various soliton molecules (SMs) with different pulse spacings and phases using composite filtering effects in a fiber laser. The manipulation of synthesizing SMs is achieved through the composite filtering functions. Our research provides a novel method for the manipulation of SM generation in a noise-driven nonlinear dissipative system, which has significant applications in the fields of pulse coding and coherent pulse detection.
APPLIED PHYSICS B-LASERS AND OPTICS
(2023)
Article
Optics
Jingxuan Sun, Zhen Liu, Yiqing Shu, Jianqing Li, Weicheng Chen
Summary: This study presents a novel method for automatically and precisely reproducing targeted soliton states in a mode-locked fiber laser using deep learning informed by spectrotemporal domain. The reproduction algorithm combines pulse information in both spectral and temporal domains, achieving successful replication of targeted solitons and advancing ultrafast laser technology.
Article
Chemistry, Analytical
Wenting Li, Jianqing Li, Di Han
Summary: In this study, a dynamic lane reversal strategy (DLRS) based on the density of congestion clusters is proposed to cope with tidal traffic. The results of simulations demonstrate that DLRS has better adaptability.
Article
Chemistry, Analytical
Penglai Guo, Huanhuan Liu, Zhitai Zhou, Jie Hu, Yuntian Wang, Xiaoling Peng, Xun Yuan, Yiqing Shu, Yingfang Zhang, Hong Dang, Guizhen Xu, Aoyan Zhang, Chenlong Xue, Jiaqi Hu, Liyang Shao, Jinna Chen, Jianqing Li, Perry Ping Shum
Summary: A fiber speckle sensor based on a tapered multimode fiber has been developed to measure liquid analyte refractive index. The sensor enhances the refractive index sensitivity and provides a resolution of 5.84 x 10(-5) over a linear response range. The results demonstrate the potential of the speckle sensor in image-based ocean-sensing applications.
Article
Computer Science, Information Systems
Kai Cui, Ruizhe Hao, Yuling Huang, Jianqing Li, Yunlin Song
Summary: This paper proposes a multi-scale convolutional neural feature extraction network (MS-CNN) for stock data, which can better extract stock trend features and make better decisions.
Article
Automation & Control Systems
Kai Fang, Tingting Wang, Lianghuai Tong, Xiaofen Fang, Yuanyuan Pan, Wei Wang, Jianqing Li
Summary: This paper proposes a security assessment method for Internet of Vehicles devices based on Microcontroller Unit (MCU) temperature, which can detect the security status of vehicles in real-time without modifying the hardware. It also introduces a Cloud-Edge-End framework to improve the execution efficiency. The experiments on Raspberry Pi 4B and Stm32 platforms demonstrate the effectiveness of the proposed method.
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2023)