다중혈관 관상동맥 환자에서 y-문합을 이용하여 양쪽 내흉동맥만을 사용한 우회술의 조기 성적

Data Mining - Concepts and Techniques.

Matrix Factorization Techniques for Recommender Systems

Click-through rate prediction is an essential task in industrial applications, such as online advertising. Recently deep learning based models have been proposed, which follow a similar Embedding&MLP paradigm. In these methods large scale sparse input features are first mapped into low dimensional embedding vectors, and then transformed into fixed-length vectors in a group-wise manner, finally concatenated together to fed into a multilayer perceptron (MLP) to learn the nonlinear relations among features. In this way, user features are compressed into a fixed-length representation vector, in regardless of what candidate ads are. The use of fixed-length vector will be a bottleneck, which brings difficulty for Embedding&MLP methods to capture user's diverse interests effectively from rich historical behaviors. In this paper, we propose a novel model: Deep Interest Network (DIN) 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. This representation vector varies over different ads, improving the expressive ability of model greatly. Besides, we develop two techniques: mini-batch aware regularization and data adaptive activation function which can help training industrial deep networks with hundreds of millions of parameters. Experiments on two public datasets as well as an Alibaba real production dataset with over 2 billion samples demonstrate the effectiveness of proposed approaches, which achieve superior performance compared with state-of-the-art methods. DIN now has been successfully deployed in the online display advertising system in Alibaba, serving the main traffic.

Deep Interest Network for Click-Through Rate Prediction

Many interesting problems in machine learning are being revisited with new deep learning tools. For graph-based semi-supervised learning, a recent important development is graph convolutional networks (GCNs), which nicely integrate local vertex features and graph topology in the convolutional layers. Although the GCN model compares favorably with other state-of-the-art methods, its mechanisms are not clear and it still requires considerable amount of labeled data for validation and model selection. In this paper, we develop deeper insights into the GCN model and address its fundamental limits. First, we show that the graph convolution of the GCN model is actually a special form of Laplacian smoothing, which is the key reason why GCNs work, but it also brings potential concerns of over-smoothing with many convolutional layers. Second, to overcome the limits of the GCN model with shallow architectures, we propose both co-training and self-training approaches to train GCNs. Our approaches significantly improve GCNs in learning with very few labels, and exempt them from requiring additional labels for validation. Extensive experiments on benchmarks have verified our theory and proposals.

https://aaai.org/ocs/index.php/AAAI/AAAI18/paper/download/16098/16553

Deeper Insights into Graph Convolutional Networks for Semi-Supervised Learning

Provided is an intelligent recommendation method, comprising: acquiring recommendation system state parameters according to multiple past historical recommended objects and the behaviors, such as the number of clicks and the number of downloads, of a user for each historical recommended object; dividing objects to be recommended into multiple levels of sets, wherein there is a subordination relationship between the various levels of sets, and each set corresponds to one selection strategy; and determining, according to the recommendation system state parameters and selection strategies for the sets, a target object to be recommended. The method is applicable to various recommendation-related application scenarios, such as application recommendation of an application market, audio/video recommendation of audio/video websites and information recommendation of an information platform. The method facilitates the improvement of the recommendation efficiency and accuracy rate.

Recommendation Method and Apparatus

A topic recommending method and device The method comprises: collecting historical operation behavior of a sample user with respect to M objects and predicting a preference value of a target user with respect to each object of the M objects according to the historical operation behavior of the sample user with respect to each object of the M objects (110); collecting category data of N recommendation objects and categorizing the N recommendation objects according to the category data of the N recommendation objects, so as to obtain X topics (120); each topic of the X topics includes at least one recommendation object of the N recommendation objects, the N recommendation objects being a portion of or all of the M objects; computing and obtaining a preference value of the target user with respect to each topic of the X topics according to a preference value of the target user with respect to a recommendation object included in each topic of the X topics (130); pushing a target topic to the target user (140), the target topic being a topic of the X topics having a preference value of the target user greater than a recommendation threshold

Topic recommending method and device

With the development of recommender systems, it becomes an increasingly common need to mix multiple item sequences from different sources. Therefore, the integrated ranking stage is proposed to be responsible for this task with re-ranking models. However, existing methods ignore the relation between the sequences, thus resulting in local optimum over the interaction session. To resolve this challenge, in this paper, we propose a new model named NFIRank (News Feed Integrated Ranking with reinforcement learning) and formulate the whole interaction session as a MDP (Markov Decision Process). Sufficient offline experiments are provided to verify the effectiveness of our model. In addition, we deployed our model on Huawei Browser and gained 1.58% improvements in CTR compared with the baseline in online A/B test. Code will be available at https://gitee.com/mindspore/models/tree/master/research/recommend/NFIRank.

Integrated Ranking for News Feed with Reinforcement Learning

As a critical task for large-scale commercial recommender systems, reranking has shown the potential of improving recommendation results by uncovering mutual influence among items. Reranking rearranges items in the initial ranking lists from the previous ranking stage to better meet users' demands. However, rather than considering the context of initial lists as most existing methods do, an ideal reranking algorithm should consider the counterfactual context -- the position and the alignment of the items in the reranked lists. In this work, we propose a novel pairwise reranking framework, Context-aware Reranking with Utility Maximization for recommendation (CRUM), which maximizes the overall utility after reranking efficiently. Specifically, we first design a utility-oriented evaluator, which applies Bi-LSTM and graph attention mechanism to estimate the listwise utility via the counterfactual context modeling. Then, under the guidance of the evaluator, we propose a pairwise reranker model to find the most suitable position for each item by swapping misplaced item pairs. Extensive experiments on two benchmark datasets and a proprietary real-world dataset demonstrate that CRUM significantly outperforms the state-of-the-art models in terms of both relevance-based metrics and utility-based metrics.

Context-aware Reranking with Utility Maximization for Recommendation

Personalized recommender systems are playing an increasingly important role as more content and services become available and users struggle to identify what might interest them. Although matrix factorization and deep learning based methods have proved effective in user preference modeling, they violate the triangle inequality and fail to capture fine-grained preference information. To tackle this, we develop a distance-based recommendation model with several novel aspects: (i) each user and item are parameterized by Gaussian distributions to capture the learning uncertainties; (ii) an adaptive margin generation scheme is proposed to generate the margins regarding different training triplets; (iii) explicit user-user/item-item similarity modeling is incorporated in the objective function. The Wasserstein distance is employed to determine preferences because it obeys the triangle inequality and can measure the distance between probabilistic distributions. Via a comparison using five real-world datasets with state-of-the-art methods, the proposed model outperforms the best existing models by 4-22% in terms of recall@K on Top-K recommendation.

Ruiming Tang

Papers

Recommendation Method and Apparatus

Topic recommending method and device

Integrated Ranking for News Feed with Reinforcement Learning

Context-aware Reranking with Utility Maximization for Recommendation

Probabilistic Metric Learning with Adaptive Margin for Top-K Recommendation