Combining semantic information with behavioral data is a crucial research area in recommender systems. A promising approach involves leveraging external knowledge to enrich behavioral-based recommender systems with abundant semantic information. However, this approach faces two primary challenges: (1) denoising raw external knowledge and (2) adapting semantic representations. To address these challenges, we propose exTernal knowledge-enhanced RecommendAtion With LLM assistance (TRAWL). This method utilizes large language models to extract relevant recommendation knowledge from raw external data and employs a contrastive learning strategy for adapter training. Experiments on public datasets and real-world online recommender systems validate the effectiveness of our approach.

Recommendation platforms commonly offer diverse content scenarios for users to interact with. Pre-training models are the most commonly used approach in recommendation systems to capture users' full-domain interests. Traditional ID-based pre-training models mainly capture user interests by leveraging collaborative signals. However, a prevalent drawback of those systems is the incapacity to handle cold-start scenarios. With the recent advent of large language models, there has been a significant increase in research efforts exploiting LLMs to extract semantic information for items. However, text-based recommendations highly rely on elaborate feature engineering and often fail to capture collaborative similarities.

Recently, cross-domain recommendation (CDR) has been widely studied in both research and industry since it can alleviate a long-standing challenge of traditional recommendation methods, i.e., data sparsity issue, by transferring the information from a relatively richer domain (termed source domain) to a sparser domain (termed target domain). To our best knowledge, most (if not all) existing CDR methods focus on transferring either the similar content information or the user preferences embedding from the source domain to the target domain. However, they fail to improve the recommendation performance in real-world recommendation scenarios where the items in the source domain are totally different from those in the target domain in terms of attributes. To solve the above issues, we analyzed the historical interactions of users from different domains in the WeChat platform, and found that if two users have similar interests (interactions) in one domain, they are very likely to have similar interests in another domain even though the items of these two domains are totally different in terms of attributes. Based on this observation, in this paper, we propose a novel model named Dual Interests-Aligned Graph Auto-Encoders (DIAGAE) by utilizing the inter-domain interest alignment of users. Besides, our proposed model DIAGAE also leverages graph decoding objectives to align intra-domain user interests, which makes the representation of two users who have similar interests in a single domain closer. Comprehensive experimental results demonstrate that our model DIAGAE outperforms state-of-the-art methods on both public benchmark datasets and online A/B tests in WeChat live-stream recommendation scenario. Our model DIAGAE now serves the major online traffic in WeChat live-streaming recommendation scenario.

Recent studies on Graph Convolutional Networks (GCNs) reveal that the initial node representations (i.e., the node representations before the first-time graph convolution) largely affect the final model performance. However, when learning the initial representation for a node, most existing work linearly combines the embeddings of node features, without considering the interactions among the features (or feature embeddings). We argue that when the node features are categorical, e.g., in many real-world applications like user profiling and recommender system, feature interactions usually carry important signals for predictive analytics. Ignoring them will result in suboptimal initial node representation and thus weaken the effectiveness of the follow-up graph convolution. In this paper, we propose a new GCN model named CatGCN, which is tailored for graph learning on categorical node features. Specifically, we integrate two ways of explicit interaction modeling into the learning of initial node representation, i.e., local interaction modeling on each pair of node features and global interaction modeling on an artificial feature graph. We then refine the enhanced initial node representations with the neighborhood aggregation-based graph convolution. We train CatGCN in an end-to-end fashion and demonstrate it on the task of node classification. Extensive experiments on three tasks of user profiling (the prediction of user age, city, and purchase level) from Tencent and Alibaba datasets validate the effectiveness of CatGCN, especially the positive effect of performing feature interaction modeling before graph convolution.

In recommender systems, some features directly affect whether an interaction would happen, making the happened interactions not necessarily indicate user preference. For instance, short videos are objectively easier to finish even though the user may not like the video. We term such feature as confounding feature , and video length is a confounding feature in video recommendation. If we fit a model on such interaction data, just as done by most data-driven recommender systems, the model will be biased to recommend short videos more, and deviate from user actual requirement. This work formulates and addresses the problem from the causal perspective. Assuming there are some factors affecting both the confounding feature and other item features, e.g., the video creator, we find the confounding feature opens a backdoor path behind user-item matching and introduces spurious correlation. To remove the effect of backdoor path, we propose a framework named Deconfounding Causal Recommendation (DCR) , which performs intervened inference with do-calculus . Nevertheless, evaluating do-calculus requires to sum over the prediction on all possible values of confounding feature, significantly increasing the time cost. To address the efficiency challenge, we further propose a mixture-of-experts (MoE) model architecture, modeling each value of confounding feature with a separate expert module. Through this way, we retain the model expressiveness with few additional costs. We demonstrate DCR on the backbone model of neural factorization machine (NFM) , showing that DCR leads to more accurate prediction of user preference with small inference time cost. We release our code at: https://github.com/zyang1580/DCR .

Recently, graph neural network (GNN) approaches have received huge interests in recommendation tasks due to their ability of learning more effective user and item representations. However, existing GNN-based recommendation models cannot support real-time recommendation where the model keeps its freshness by continuously training the streaming data that users produced, leading to negative impact on recommendation performance. To fully support graph-enhanced large-scale recommendation in real-time scenarios, a deep graph learning system is required to dynamically store the streaming data as a graph structure and enable the development of any GNN model incorporated with the capabilities of real-time training and online inference. However, such requirements rule out existing deep graph learning solutions. In this paper, we propose a new deep graph learning system called PlatoGL, where (1) an effective block-based graph storage is designed with non-trivial insertion/deletion mechanism for updating the graph topology in-milliseconds, (2) a non-trivial multi-blocks neighbour sampling method is proposed for efficient graph query, and (3) a cache technique is exploited to improve the storage stability. We have deployed PlatoGL in Wechat, and leveraged its capability in various content recommendation scenarios including live-streaming, article and micro-video. Comprehensive experiments on both deployment performance and benchmark performance~(w.r.t. its key features) demonstrate its effectiveness and scalability. One real-time GNN-based model, developed with PlatoGL, now serves the major online traffic in WeChat live-streaming recommendation scenario.

Sequential recommender systems (SRSs) have become a research hotspot recently due to its powerful ability in capturing users' dynamic preferences. The key idea behind SRSs is to model the sequential dependencies over the user-item interactions. However, we argue that users' preferences are not only determined by their view or purchase items but also affected by the item-providers with which users have interacted. For instance, in a short-video scenario, a user may click on a video because he/she is attracted to either the video content or simply the video-providers as the vloggers are his/her idols. Motivated by the above observations, in this paper, we propose IPSRec, a novel Item-Provider co-learning framework for Sequential Recommendation. Specifically, we propose two representation learning methods (single-steam and cross-stream) to learn comprehensive item and user representations based on the user's historical item sequence and provider sequence. Then, contrastive learning is employed to further enhance the user embeddings in a self-supervised manner, which treats the representations of a specific user learned from the item side as well as the item-provider side as the positive pair and treats the representations of different users in the batch as the negative samples. Extensive experiments on three real-world SRS datasets demonstrate that IPSRec achieves substantially better results than the strong competitors. For reproducibility, our code and data are available at https://github.com/siat-nlp/IPSRec.

Recommender system usually faces popularity bias issues: from the data perspective, items exhibit uneven (usually long-tail) distribution on the interaction frequency; from the method perspective, collaborative filtering methods are prone to amplify the bias by over-recommending popular items. It is undoubtedly critical to consider popularity bias in recommender systems, and existing work mainly eliminates the bias effect with propensity-based unbiased learning or causal embeddings. However, we argue that not all biases in the data are bad, \ie some items demonstrate higher popularity because of their better intrinsic quality. Blindly pursuing unbiased learning may remove the beneficial patterns in the data, degrading the recommendation accuracy and user satisfaction. This work studies an unexplored problem in recommendation --- how to leverage popularity bias to improve the recommendation accuracy. The key lies in two aspects: how to remove the bad impact of popularity bias during training, and how to inject the desired popularity bias in the inference stage that generates top-K recommendations. This questions the causal mechanism of the recommendation generation process. Along this line, we find that item popularity plays the role ofconfounder between the exposed items and the observed interactions, causing the bad effect of bias amplification. To achieve our goal, we propose a new training and inference paradigm for recommendation named Popularity-bias Deconfounding and Adjusting (PDA). It removes the confounding popularity bias in model training and adjusts the recommendation score with desired popularity bias via causal intervention. We demonstrate the new paradigm on the latent factor model and perform extensive experiments on three real-world datasets from Kwai, Douban, and Tencent. Empirical studies validate that the deconfounded training is helpful to discover user real interests and the inference adjustment with popularity bias could further improve the recommendation accuracy. We release our code at https://github.com/zyang1580/PDA.

Relationships in online social networks often imply social connections in real life. An accurate understanding of relationship types benefits many applications, e.g. social advertising and recommendation. Some recent attempts have been proposed to classify user relationships into predefined types with the help of pre-labeled relationships or abundant interaction features on relationships. Unfortunately, both relationship feature data and label data are very sparse in real social platforms like WeChat, rendering existing methods inapplicable. In this paper, we present an in-depth analysis of WeChat relationships to identify the major challenges for the relationship classification task. To tackle the challenges, we propose a Local Community-based Edge Classification (LoCEC) framework that classifies user relationships in a social network into real-world social connection types. LoCEC enforces a three-phase processing, namely local community detection, community classification and relationship classification, to address the sparsity issue of relationship features and relationship labels. Moreover, LoCEC is designed to handle large-scale networks by allowing parallel and distributed processing. We conduct extensive experiments on the real-world WeChat network with hundreds of billions of edges to validate the effectiveness and efficiency of LoCEC.

The diffusion of rumors is a major concern for web users. Limiting the spread of rumor on social networks has become an important task. One approach is to identify nodes to start a truth campaign such that when users are aware of the truth, they would not believe or propagate the rumor. However, existing works do not take into account the delays of information diffusion or the time point beyond which propagation of misinformation is no longer critical. In this paper, we consider a more realistic situation where information is propagated with delays and the goal is to reduce the number of rumor-infected users before a deadline. We call this the Temporal Influence Blocking (TIB) problem. We propose a two-phase solution called TIB-Solver to select k nodes to start a truth campaign such that the number of users reached by a rumor is minimized. Experiments show that the proposed TIBSolver outperforms the state-of-the-art algorithms in terms of both effectiveness and efficiency.

Influence maximization (IM) problem asks for a set of k nodes in a given graph G, such that it can reach the largest expected number of remaining nodes in G. Existing methods have either considered that the influence be targeted to meet certain deadline constraint, or be restricted to specific geographical region. However, if an event organizer wants to disseminate some event information on a social platform, s/he would want to select a set of users who can influence the most number of people within the neighborhood of the event location, and this influence should occur before the event takes place. Considering the location and deadline independently may lead to a less than optimal set of users. In this paper, we formalize the problem targeted influence maximization in social networks. We adopt a login model where each user is associated with a login probability and he can be influenced by his neighbors only when he is online. We develop a sampling based algorithm that returns a (1-1/e-ε)-approximate solution, as well as an efficient heuristic algorithm that focuses on nodes close to the target location. Experiments on real-world social network datasets demonstrate the effectiveness and efficiency of our proposed method.

The theory of brokerage in sociology suggests if contacts between two parties are enabled through a third party, the latter occupies a strategic position of controlling information flows. Such individuals are called brokers and they play a key role in disseminating information. However, there is no systematic approach to identify brokers in online social networks. In this paper, we formally define the problem of detecting top-$k$ brokers given a social network and show that it is NP-hard. We develop a heuristic algorithm to find these brokers based on the weak tie theory. In order to handle the dynamic nature of online social networks, we design incremental algorithms: WeakTie-Local for unidirectional networks and WeakTie-Bi for bidirectional networks. We use two real world datasets, DBLP and Twitter, to evaluate the proposed methods. We also demonstrate how the detected brokers are useful in diffusing information across communities and propagating tweets to reach more distinct users.

Locating nodes to immunize in computer/social networks to control the spread of virus or rumors has become an important problem. In real world contagions, nodes may get infected by external sources when the propagation is underway. While most studies formalize the problem in a setting where contagion starts at one time point, we model a more realistic situation where there are likely to be many breakouts of contagions over a time window. We call this the node immunization over infectious period (NIIP) problem. We show that the NIIP problem is NP-hard and remains so even in directed acyclic graphs. We propose a NIIP algorithm to select $k$ nodes to immunize over a time period. Simulation is performed to estimate a good distribution of $k$ over the time period. For each time point, the NIIP algorithm will make decisions which nodes to immunize given the estimated value of $k$ for that time point. Experiments show that the proposed NIIP algorithm outperform the state-of-the-art algorithms in terms of both effectiveness and efficiency.