A CTR prediction model based on session interest

Review Engineering, Electrical & Electronic

Data Sampling Strategies for Click Fraud Detection Using Imbalanced User Click Data of Online Advertising: An Empirical Review

Deepti Sisodia et al.

Summary: This study conducted an empirical evaluation of fraudster classification in online advertising, finding that the combination of adaptive synthetic sampling (ADASYN) oversampling and gradient tree boosting (GTB) model performed the best.

IETE TECHNICAL REVIEW (2022)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Modeling user micro-behaviors and original interest via Adaptive Multi-Attention Network for session-based recommendation

Jingjing Qiao et al.

Summary: This article introduces a novel model called AMAN (Adaptive Multi-Attention Network) for session-based recommendation. AMAN considers both user micro-behaviors and original interest, and has proven to outperform the state-of-the-art baselines in experimental results.

KNOWLEDGE-BASED SYSTEMS (2022)

添加到收藏夹

Review Computer Science, Interdisciplinary Applications

Computer Vision Techniques in Construction: A Critical Review

Shuyuan Xu et al.

Summary: Computer vision has gained significant interest in various research areas, with the architecture, engineering, construction, and facility management sector being one of the most intensive fields using vision-based systems/methods. Research indicates the potential of computer vision in assisting on-site managerial tasks in construction sites, while also presenting challenges.

ARCHIVES OF COMPUTATIONAL METHODS IN ENGINEERING (2021)

添加到收藏夹

Article Physics, Multidisciplinary

Personal Interest Attention Graph Neural Networks for Session-Based Recommendation

Xiangde Zhang et al.

Summary: The PIA-GNN method proposes a personalized interest attention graph neural network that utilizes PGNN to capture complex transitions between items and activate users' interest in different items adaptively. Additionally, a self-attention layer is used to capture long-term dependencies between items and users' long-term preferences. The experimental results show that PIA-GNN outperforms existing personalized session-aware recommendation methods.

ENTROPY (2021)

添加到收藏夹

Article Construction & Building Technology

Temporal image analytics for abnormal construction activity identification

Zi-Hao Lin et al.

Summary: This paper proposes a method for analyzing abnormal activities on construction jobsites through consecutive image sequences, including object detection, object tracking, action recognition, and operational analysis. The experimental results show that the method can effectively identify irregular operations and provide field managers with the opportunity for further investigation and adjustments.

AUTOMATION IN CONSTRUCTION (2021)

添加到收藏夹

Article Engineering, Multidisciplinary

A Hybrid Neural Network Architecture to Predict Online Advertising Click-Through Rate Behaviors in Social Networks

Renjie Zhou et al.

Summary: This study introduces a new hybrid neural network model, DGRU, which integrates DeepFM and GRU components, with DeepFM performing autonomic feature combination and GRU modeling user interests and evolutions. The GRU component is fed with a series of 1 and 0 representing user click behaviors, distinguishing between user likes and dislikes. Experiments demonstrate that the proposed model outperforms existing models in CTR prediction performance and robustness.

IEEE TRANSACTIONS ON NETWORK SCIENCE AND ENGINEERING (2021)

添加到收藏夹

Article Computer Science, Theory & Methods

A Survey on Conversational Recommender Systems

Dietmar Jannach et al.

Summary: Recommender systems are software applications that assist users in finding items of interest in situations of information overload, while conversational recommender systems offer a richer set of interactions. These interactions can help improve preference elicitation and allow users to ask questions and provide feedback.

ACM COMPUTING SURVEYS (2021)

添加到收藏夹

Article Computer Science, Interdisciplinary Applications

A new design of evolutionary hybrid optimization of SVR model in predicting the blast-induced ground vibration

Wusi Chen et al.

Summary: This study examines the suitability of hybridizing different optimization algorithms with data-driven models to predict ground vibration, and compares the performance of various models through a case study, with the MFA-SVR model being the most accurate.

ENGINEERING WITH COMPUTERS (2021)

添加到收藏夹

Article Computer Science, Artificial Intelligence