21 March 2024 in IEEE Transactions on Services Computing
Recommendation systems greatly improve user experience with personalized suggestions. Much research utilizes deep learning to extract user interest features, yet often neglects their real-time and dynamic aspects. This paper proposes a joint learning-based recommendation model (JLS-Rec) for e-commerce platforms, integrating both short-term and long-term user interests. First, for users' long-term interests, the paper suggests mining more detailed interests from user behavior sequences. This method decouples the behavior sequence in horizontal and vertical directions using a Convolutional Neural Network to learn the user's long-term interests. Second, for users' short-term interests, the JLS-Rec method learns feature transformations on users' recent behavior sequences by stacking multiple self-attention mechanisms, resulting in dynamic representations of the user's short-term interests at the current stage. Finally, based on the principle of prioritizing short-term memory without neglecting long-term interests, the paper proposes a joint learning framework with dual embeddings to balance the two characteristics of user long-term and short-term interests. This framework generates accurate recommendation results while utilizing these two interest features to predict user feedback on products. The experimental results demonstrate that the model effectively mines the long-term and short-term interest information of users in the features, thereby improving the recommendation accuracy of e-commerce platforms.
9 September 2025 in IEEE Transactions on Emerging Topics in Computational Intelligence
Organizations often deploy marketing campaigns on e-commerce service platforms to make profit. This paper presents a model for predicting the monetization rate of marketing campaigns run on e-commerce service platforms based on hidden information. Hidden information consists of two parts: user–campaign potential relations and implicit features. The relation data between entities (users and marketing campaigns) on the platform is complex and diverse, a new method called II2vec is used to represent the implicit information of user–campaign relations as vectors. To address the problem of incomplete true information of raw feature descriptions, a compressed interaction network (CIN) is used to extract valid implicit features and discover user-derived behaviors to further improve the model's learning performance. The proposed monetization prediction model (dynamic excitation convolutional neural network), which fuses the previous two kinds of hidden information, is able to analyze the force of campaigns on users while capturing local time-space features well, realistically depicts the psychological state of users at the edge of the purchase period, and dynamically simulates user's future behavior. The experimental results demonstrate that the model can both accurately predict the monetization rate of marketing campaigns run on e-commerce service platforms and reflect the multiple influence factors that drive the user purchase behavior.