User Behavior Retrieval for Click-Through Rate Prediction

@article{Qin2020UserBR,
  title={User Behavior Retrieval for Click-Through Rate Prediction},
  author={Jiarui Qin and Weinan Zhang and Xin Wu and Jiarui Jin and Yuchen Fang and Yong Yu},
  journal={Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval},
  year={2020}
}
  • Jiarui Qin, Weinan Zhang, Yong Yu
  • Published 28 May 2020
  • Computer Science
  • Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval
Click-through rate (CTR) prediction plays a key role in modern online personalization services. In practice, it is necessary to capture user's drifting interests by modeling sequential user behaviors to build an accurate CTR prediction model. However, as the users accumulate more and more behavioral data on the platforms, it becomes non-trivial for the sequential models to make use of the whole behavior history of each user. First, directly feeding the long behavior sequence will make online… 
End-to-End User Behavior Retrieval in Click-Through RatePrediction Model
TLDR
A localitysensitive hashing (LSH) method called ETA (End- to-end Target Attention) which can greatly reduce the training and inference cost and make the end-to-end training with long-term user behavior sequence possible and confirm the effectiveness of the model.
Rethinking Lifelong Sequential Recommendation with Incremental Multi-Interest Attention
TLDR
A novel lifelong incrementalmulti-interest self attention based sequential recommendation model, namely LimaRec, which benefits from the carefully designed self-attention to identify relevant information from users’ behavior sequences with different interests and is still able to incrementally update users' representations for online inference, similarly to memory network based approaches.
Adversarial Filtering Modeling on Long-term User Behavior Sequences for Click-Through Rate Prediction
TLDR
A novel Adversarial Filtering Model (ADFM) is proposed to model long-term user behavior that uses a hierarchical aggregation representation to compress raw behavior sequence and then learns to remove useless behavior information with an adversarial filtering mechanism.
Relation‐Level User Behavior Modeling for Click‐Through Rate Prediction
TLDR
This paper decomposes the long item sequence into a group of extremely short sequences (item pairs) and conducts relational reasoning on them, which consists of two parts, which are designed for capturing various transitional patterns in the behavior sequences.
Learn over Past, Evolve for Future: Search-based Time-aware Recommendation with Sequential Behavior Data
TLDR
A novel framework named Search-based Time-Aware Recommendation (STARec), which captures the evolving demands of users over time through a unified search-based time-aware model and develops a novel label trick that uses the previous labels as the input to better capture the user’s browsing pattern.
Click-Through Rate Prediction with Multi-Modal Hypergraphs
TLDR
A Hypergraph Click-Through Rate prediction framework built upon the hyperedge notion of hypergraph neural networks, which can yield modal-specific representations of users and micro-videos to better capture user preferences and significantly outperforms various state-of-the-art methods.
Dual Graph enhanced Embedding Neural Network for CTR Prediction
TLDR
A novel module named Dual Graph enhanced Embedding, which is compatible with various CTR prediction models to alleviate the feature sparsity and behavior sparsity problems, and the experimental results show that the proposed DG-ENN significantly outperforms state-of-the-art CTR Prediction models.
RLNF: Reinforcement Learning based Noise Filtering for Click-Through Rate Prediction
TLDR
A reinforcement learning based noise filtering approach, dubbed RLNF, which employs a noise filter to select effective negative samples that can be used to enhance the CTR prediction model, and meanwhile the effectiveness of the noise filter can be enhanced through reinforcement learning using the performance of CTR prediction models as reward.
ReLoop: A Self-Correction Continual Learning Loop for Recommender Systems
TLDR
This paper attempts to build a self-correction learning loop (dubbed ReLoop) for recommender systems that enables a continual self-Correction process in the long run and thus is expected to obtain better performance over existing training strategies.
MISS: Multi-Interest Self-Supervised Learning Framework for Click-Through Rate Prediction
TLDR
A novel Multi-Interest Self-Supervised learning (MISS) framework which enhances the feature embeddings with interest-level self-supervision signals and can be used as an “plug-in” component with existing CTR prediction models and further boost their performances.
...
1
2
3
...

References

SHOWING 1-10 OF 51 REFERENCES
Practice on Long Sequential User Behavior Modeling for Click-Through Rate Prediction
TLDR
This paper faces directly the challenge of long sequential user behavior modeling and proposes a novel memory-based architecture named MIMN (Multi-channel user Interest Memory Network) to capture user interests from long sequential behavior data, achieving superior performance over state-of-the-art models.
Lifelong Sequential Modeling with Personalized Memorization for User Response Prediction
TLDR
A Hierarchical Periodic Memory Network for lifelong sequential modeling with personalized memorization of sequential patterns for each user is proposed and adopts a hierarchical and periodical updating mechanism to capture multi-scale sequential patterns of user interests while supporting the evolving user behavior logs.
Deep Interest Evolution Network for Click-Through Rate Prediction
TLDR
This paper proposes a novel model, named Deep Interest Evolution Network~(DIEN), for CTR prediction, which significantly outperforms the state-of-the-art solutions and design interest extractor layer to capture temporal interests from history behavior sequence.
Self-Attentive Sequential Recommendation
TLDR
Extensive empirical studies show that the proposed self-attention based sequential model (SASRec) outperforms various state-of-the-art sequential models (including MC/CNN/RNN-based approaches) on both sparse and dense datasets.
Deep Interest Network for Click-Through Rate Prediction
TLDR
A novel model: Deep Interest Network (DIN) is proposed which tackles this challenge by designing a local activation unit to adaptively learn the representation of user interests from historical behaviors with respect to a certain ad.
Fusing Similarity Models with Markov Chains for Sparse Sequential Recommendation
TLDR
Fossil is proposed, a similarity-based method that outperforms alternative algorithms, especially on sparse datasets, and qualitatively that it captures personalized dynamics and is able to make meaningful recommendations.
Sequential Recommendation with User Memory Networks
TLDR
A memory-augmented neural network (MANN) integrated with the insights of collaborative filtering for recommendation is designed, which store and update users» historical records explicitly, which enhances the expressiveness of the model.
Dual Sequential Prediction Models Linking Sequential Recommendation and Information Dissemination
TLDR
This paper proposes dual sequential prediction models that unify these two thinking paradigms and shows that the dual models can better distinguish false negative samples and true negative samples compared with single sequential recommendation or information dissemination models.
Sequential Recommendation with Dual Side Neighbor-based Collaborative Relation Modeling
TLDR
This work proposes Sequential Collaborative Recommender (SCoRe) which effectively mines high-order collaborative information using cross-neighbor relation modeling and, additionally utilizes both user-side and item-side historical sequences to better capture user and item dynamics.
Context-Aware Sequential Recommendation
TLDR
Experimental results show that the proposed CA-RNN model yields significant improvements over state-of-the-art sequential recommendation methods and context-aware recommendation methods on two public datasets, i.e., the Taobao dataset and the Movielens-1M dataset.
...
1
2
3
4
5
...