Search engines, Recommender systems, and Online advertising are playing fundamental roles in modern web and mobile applications. In these information systems, the most significant component is the ranking system, which selects a list of items likely to interest a user from billions of candidate items. At its core, Deep learning to Rank (DLTR) has become indispensable for building high-performance ranking models, driving significant gains in user engagement and business growth. In this article, firstly, we outline the key problems and challenges in industrial-scale ranking systems. Secondly, we provide a comprehensive review of deep learning models deployed across multiple stages of the industrial ranking pipeline, including matching, pre-ranking, fine-grained ranking, post-ranking, and relevance-ranking. Finally, we explore novel perspectives for future research, such as leveraging Large Language Models (LLMs). The papers discussed in this survey are listed in https://github.com/guyulongcs/Awesome-Deep-Learning-Papers-for-Search-Recommendation-Advertising .

User feedback data (e.g., clicks, dwell time in the product detail page) have been incorporated in the training process of many ranking models for better performance. Such approaches are widely used in many ranking applications, including search and recommendation. Recently, the inherent biases in user feedback data have been studied, which indicates how the users’ behaviors can be affected by factors other than relevancy. By identifying and removing these biases, the ranking models can be further improved. Researchers have developed a variety of debiasing methods on different bias factors. Most of them only focus on one type of bias and pay little attention to different types of bias from a unified perspective. In this paper, we conduct a comprehensive study of bias focusing on the application of ranking problems in recommender systems which is highly important for the research of web intelligence. Then, we share our experiences derived from designing and optimizing unbiased models to improve feeds recommendation. To uncover the effects of biases and achieve better ranking performance, we propose several unbiased models and compare with state-of-the-art models. We conduct extensive offline experiments on real datasets and validate the effectiveness of our method by performing online A/B testing in a real-world recommender system.

Modeling user's historical feedback is essential for Click-Through Rate Prediction in personalized search and recommendation. Existing methods usually only model users' positive feedback information such as click sequences which neglects the context information of the feedback. In this paper, we propose a new perspective for context-aware users' behavior modeling by including the whole page-wisely exposed products and the corresponding feedback as contextualized page-wise feedback sequence. The intra-page context information and inter-page interest evolution can be captured to learn more specific user preference. We design a novel neural ranking model RACP(Recurrent Attention over Contextualized Page sequence), which utilizes page-context aware attention to model the intra-page context. A recurrent attention process is used to model the cross-page interest convergence evolution as denoising the interest in the previous pages. Experiments on public and real-world industrial datasets verify our model's effectiveness.

Event sequence, where each event is associated with a marker and a timestamp, is increasingly ubiquitous in various applications. Accordingly, event forecasting emerges to be a crucial problem, which aims to predict the next event based on the historical sequence. In this paper, we propose ANPP, an Attentive Neural Point Processes framework to solve this problem. In comparison with state-of-the-art methods like recurrent marked temporal point processes, ANPP leverages the time-aware self-attention mechanism to explicitly model the influence between every pair of historical events, resulting in more accurate predictions of events and better interpretation ability. Extensive experiments on one synthetic and four real-world datasets demonstrate that ANPP can achieve significant performance gains against state-of-the-art methods for predictions of both timings and markers. To facilitate future research, we release the codes and datasets at https://github.com/guyulongcs/AAAI2021\_ANPP.

Multi-scenario Learning to Rank is essential for Recommender Systems, Search Engines and Online Advertising in e-commerce portals where the ranking models are usually applied in many scenarios. However, existing works mainly focus on learning the ranking model for a single scenario, and pay less attention to learning ranking models for multiple scenarios. We identify two practical challenges in industrial multi-scenario ranking systems: (1) The Feedback Loop problem that the model is always trained on the items chosen by the ranker itself. (2) Insufficient training data for small and new scenarios. To address the above issues, we present ZEUS, a novel framework that learns a Zoo of ranking modEls for mUltiple Scenarios based on pre-training on users' spontaneous behaviors (e.g. queries which are directly searched in the search box and not recommended by the ranking system). ZEUS decomposes the training process into two stages: self-supervised learning based pre-training and fine-tuning. Firstly, ZEUS performs self-supervised learning on users' spontaneous behaviors and generates a pre-trained model. Secondly, ZEUS fine-tunes the pre-trained model on users' implicit feedback in multiple scenarios. Extensive experiments on Alibaba's production dataset demonstrate the effectiveness of ZEUS, which significantly outperforms state-of-the-art methods. ZEUS averagely achieves 6.0%, 9.7%, 11.7% improvement in CTR, CVR and GMV respectively than state-of-the-art method.

Recommender Systems have been playing essential roles in e-commerce portals. Existing recommendation algorithms usually learn the ranking scores of items by optimizing a single task (e.g. Click-through rate prediction) based on users' historical click sequences, but they generally pay few attention to simultaneously modeling users' multiple types of behaviors or jointly optimize multiple objectives (e.g. both Click-through rate and Conversion rate), which are both vital for e-commerce sites. In this paper, we argue that it is crucial to formulate users' different interests based on multiple types of behaviors and perform multi-task learning for significant improvement in multiple objectives simultaneously. We propose Deep Multifaceted Transformers (DMT), a novel framework that can model users' multiple types of behavior sequences simultaneously with multiple Transformers. It utilizes Multi-gate Mixture-of-Experts to optimize multiple objectives. Besides, it exploits unbiased learning to reduce the selection bias in the training data. Experiments on JD real production dataset demonstrate the effectiveness of DMT, which significantly outperforms state-of-art methods. DMT has been successfully deployed to serve the main traffic in the commercial Recommender System in JD.com. To facilitate future research, we release the codes and datasets at https://github.com/guyulongcs/CIKM2020_DMT.

Inferring substitutable and complementary items is an important and fundamental concern for recommendation in e-commerce websites. However, the item relationships in real-world are usually heterogeneous, posing great challenges to conventional methods that can only deal with homogeneous relationships. More specifically, for this problem, there is a lack of in-depth investigation on 1) decoupling item semantics for modeling heterogeneous item relationships, and at the same time, 2) incorporating mutual influence between different relationships. To fill this gap, we propose a novel solution, namely Decoupled Graph Convolutional Network (DecGCN), to solve the problem of inferring substitutable and complementary items. DecGCN is designed to model item substitutability and complementarity in separated embedding spaces, and is equipped with a two-step integration scheme,where inherent influences between 1) different graph structures and 2) different item semantics are captured. Our experiments on three real-world datasets demonstrate that DecGCN is more effective than the state-of-the-art baselines for the problem at hand. We also conduct offline and online A/B tests on large-scale industrial data, where the results show that DecGCN is effective to be deployed in real-world applications. We release the codes at https://github.com/liuyiding1993/CIKM2020_DecGCN.

In this paper, we study collaborative filtering in an interactive setting, in which the recommender agents iterate between making recommendations and updating the user profile based on the interactive feedback. The most challenging problem in this scenario is how to suggest items when the user profile has not been well established, \ie recommend for cold-start users or warm-start users with taste drifting. Existing approaches either rely on overly pessimistic linear exploration strategy or adopt meta-learning based algorithms in a full exploitation way. In this work, to quickly catch up with the user's interests, we proposed to represent the exploration policy with a neural network and directly learn it from the feedback data. Specifically, the exploration policy is encoded in the weights of multi-channel stacked self-attention neural networks and trained with efficient Q-learning by maximizing users' overall satisfaction in the recommender systems. The key insight is that the satisfied recommendations triggered by the exploration recommendation can be viewed as the exploration bonus (delayed reward) for its contribution on improving the quality of the user profile. Therefore, the proposed exploration policy, to balance between learning the user profile and making accurate recommendations, can be directly optimized by maximizing users' long-term satisfaction with reinforcement learning. Extensive experiments and analysis conducted on three benchmark collaborative filtering datasets have demonstrated the advantage of our method over state-of-the-art methods.

Hierarchical user profiling that aims to model users' real-time interests in different granularity is an essential issue for personalized recommendations in E-commerce. On one hand, items (i.e. products) are usually organized hierarchically in categories, and correspondingly users' interests are naturally hierarchical on different granularity of items and categories. On the other hand, multiple granularity oriented recommendations become very popular in E-commerce sites, which require hierarchical user profiling in different granularity as well. In this paper, we propose HUP, a Hierarchical User Profiling framework to solve the hierarchical user profiling problem in E-commerce recommender systems. In HUP, we provide a Pyramid Recurrent Neural Networks, equipped with Behavior-LSTM to formulate users' hierarchical real-time interests at multiple scales. Furthermore, instead of simply utilizing users' item-level behaviors (e.g., ratings or clicks) in conventional methods, HUP harvests the sequential information of users' temporal finely-granular interactions (micro-behaviors, e.g., clicks on components of items like pictures or comments, browses with navigation of the search engines or recommendations) for modeling. Extensive experiments on two real-world E-commerce datasets demonstrate the significant performance gains of the HUP against state-of-the-art methods for the hierarchical user profiling and recommendation problems. We release the codes and datasets at https://github.com/guyulongcs/WSDM2020_HUP.

Aiming to represent user characteristics and personal interests, the task of user profiling is playing an increasingly important role for many real-world applications, e.g., e-commerce and social networks platforms. By exploiting the data like texts and user behaviors, most existing solutions address user profiling as a classification task, where each user is formulated as an individual data instance. Nevertheless, a user's profile is not only reflected from her/his affiliated data, but also can be inferred from other users, e.g., the users that have similar co-purchase behaviors in e-commerce, the friends in social networks, etc. In this paper, we approach user profiling in a semi-supervised manner, developing a generic solution based on heterogeneous graph learning. On the graph, nodes represent the entities of interest (e.g., users, items, attributes of items, etc.), and edges represent the interactions between entities. Our heterogeneous graph attention networks (HGAT) method learns the representation for each entity by accounting for the graph structure, and exploits the attention mechanism to discriminate the importance of each neighbor entity. Through such a learning scheme, HGAT can leverage both unsupervised information and limited labels of users to build the predictor. Extensive experiments on a real-world e-commerce dataset verify the effectiveness and rationality of our HGAT for user profiling.

Social Networks have experienced increased popularity and rapid growth in recent years. Recommendation is significant for users due to the extremely large amount of information in Social Networks. Most existing recommender systems rely on collaborative filtering techniques which focus on recommending the most relevant items to users based on past rating information of users or items. In Social Networks, the cold-start and data sparsity problems are very serious because new users and items are growing rapidly. Taking the Event Recommendation problem in Event-Based Social Networks as a scenario, many events are newly created and have few feedbacks. Existed collaborative filtering based methods will fail for Social Recommendation due to these problems. Therefore, a more sophisticated recommendation mechanism that can efficiently combine various contextual information to further improve recommendation quality is desired. In this paper, we propose a Context Aware Matrix Factorization model called CAMF which models implicit feedbacks and various contextual information simultaneously for Social Recommendation. Specifically, CAMF is a unified model that combines the Matrix Factorization model which models implicit feedbacks with the Linear Contextual Features model which models explicit contextual features. Extensive experiments on a large real-world dataset demonstrate that the CAMF model significantly outperforms state-of-the-art methods by 12.7% in terms of accuracy for the Event Recommendation problem.

Understanding Activities of Human Daily Life is a fundamental and essential AI problem for Pervasive Computing and Human-Computer Interaction. Activity Sensing has attracted enormous research on activity recognition from mobile sensor data. However, there are two challenging problems: There is no standard taxonomy of activities and there is a lack of research on sensing high level activities. To this end, firstly, we built AHDL, the first knowledge base of Activities of Human Daily Life in this planet leveraging a large time use surveys. AHDL not only has a taxonomy of activities but also has common sense knowledge of these activities. Secondly, we designed ActivitySensor, a Conditional Random Fields based Sensor for sensing high level activities in AHDL. To be specific, ActivitySensor performs activity sensing using Conditional Random Fields model by combining contextual signals (time, location, previous activity and related person) and demographical signals. Extensive experiments demonstrated that ActivitySensor can improve the accuracy of activity recognition about 15% comparing to state-of-the-art methods on the same dataset. What's more, we revealed that ActivitySensor can predict what will you do next with high accuracy.

The rapid spread of smartphones has led to the increasing popularity of Location-Based Social Networks(LBSNs) like Foursquare, Gowalla, Facebook Places and so on where users can publish information about their current location. In LBSNs, identifying home locations of users is very important for various applications like effective location-based advertisement and recommendation. However, this problem is rather challenging because the location information in LBSNs is sparse and noisy: Only a small percentage of users share their home location information due to privacy concerns; users may check in at diverse places far from their home and make friends far away; many users even do not have any check-in information. In this paper, we propose a trust-based influence model, named as TSU to solve the problem. To be specific, TSU is a Trust-based unified probabilistic model that models edges in LBSNs based on signals from Social relationship data(social friendship, social trust) and User-centric data(check-in data) in LBSN. We proposed a Home Location Identification method based on TSU model and evaluate it on a large real-world LBSNs dataset. Extensive experiments demonstrate that our method significantly outperforms state-of-the-art methods.

The rapid spread of mobile internet and location-acquisition technologies have led to the increasing popularity of Location-Based Social Networks(LBSNs). Users in LBSNs can share their life by checking in at various venues at any time. In LBSNs, identifying home locations of users is significant for effective location-based services like personalized search, targeted advertisement, local recommendation and so on. In this paper, we propose a Home Location Global Positioning System called HLGPS to tackle with the home location identification problem in LBSNs. Firstly, HLGPS uses an influence model named as IME to model edges in LBSNs. Then HLGPS uses a global iteration algorithm based on IME model to position home location of users so that the joint probability of generating all the edges in LBSNs is maximum. Extensive experiments on a large real-world LBSN dataset demonstrate that HLGPS significantly outperforms state-of-the-art methods by 14.7%.

Enormous efforts of human volunteers have made Wikipedia become a treasure of textual knowledge. Relation extraction that aims at extracting structured knowledge in the unstructured texts in Wikipedia is an appealing but quite challenging problem because it's hard for machines to understand plain texts. Existing methods are not effective enough because they understand relation types in textual level without exploiting knowledge behind plain texts. In this paper, we propose a novel framework called Athena 2.0 leveraging Semantic Patterns which are patterns that can understand relation types in semantic level to solve this problem. Extensive experiments show that Athena 2.0 significantly outperforms existing methods.

Event-based Social Networks(EBSNs) which combine online interactions and offline events among users have experienced increased popularity and rapid growth recently. In EBSNs, event recommendation is significant for users due to the extremely large amount of events. However, the event recommendation problem is rather challenging because it faces a serious cold-start problem: Events have short life time and new events are registered by only a few users. What's more, there are only implicit feedback information. Existing approaches like collaborative filtering methods are not suitable for this scenario. In this paper, we propose a Context Aware Matrix Factorization model called AlphaMF to tackle with the problem. Specifically, AlphaMF is a unified model that combines the Matrix Factorization model which models implicit feedbacks with the Linear contextual features model which models explicit contextual features. Extensive experiments on a large real-world EBSN dataset demonstrate that the AlphaMF model significantly outperforms state-of-the-art methods by 11%.

The development of genetic services within healthcare systems is a global phenomenon that raises challenges for managing genetic information. This paper describes an ongoing qualitative study to collect stakeholder perspectives of New Zealand (NZ) genetic services concerning genetic information management. We are conducting semi-structured interviews to build an understanding of their experiences, expectations, and concerns. The data analysis takes a general inductive approach with an analytic comparison strategy and evaluation research techniques. This study draws on past social and health science theories, on our experience in the NZ genetics context, and on emerging issues in the domain. The study result will provide deeper insights on how the genetic service system works and where it should go. The project deliverables will include a structured synthesis of stakeholder requirements, NZ genetic information management principles and goals.

Although genetic services have potential value for health care and disease control, there is no systematic solution for knowledge capture in Medical Genetic Testing (MGT) Laboratories. This paper addresses such knowledge management (KM) technology weakness in genetics domain, then proposes an information system (IS) architecture to establish process automation and content management of the distributed workflow of knowledge generation and knowledge management (KG&KM) during MGT result interpretation. The presented IS will validate the interpretation decision by using information systems/information technologies (IS/IT), esp. KM tools, such as workflow management system (WfMS), search engine and groupware. Once developed and implemented, our integrated system will significantly improve MGT lab researchers' KG&KM performance through increasing knowledge capture, improving documentation quality and maintaining (if not improving) users' information satisfaction