Yonghui Xu

Nanyang Technological University | ntu

Mining potential and valuable medical knowledge from massive medical data to support clinical decision-making has become an important research field. Personalized medicine recommendation is an important research direction in this field, aiming to recommend the most suitable medicines for each patient according to the health status of the patient. P...

Medicine package recommendation aims to assist doctors in clinical decision-making by recommending appropriate packages of medicines for patients. Current methods model this task as a munderlineti-label classification or sequence generation problem, focusing on learning relationships between individual medicines and other medical entities. However,...

Electronic health record (EHR) data is crucial in providing comprehensive historical disease information for patients and is frequently utilized in health event prediction. However, current deep learning models that rely on EHR data encounter significant challenges. These include inadequate exploration of higher‐order relationships among diseases,...

Federated learning has recently been applied to recommendation systems to protect user privacy. In federated learning settings, recommendation systems can train recommendation models by collecting the intermediate parameters instead of the real user data, which greatly enhances user privacy. In addition, federated recommendation systems (FedRSs) ca...

Accurately estimating packages’ arrival time in e-commerce can enhance users’ shopping experience and improve the placement rate of products. This problem is often formalized as an Origin-Destination (OD)-based ETA ( i.e. estimated time of arrival) prediction task, where the delivery time is estimated mainly based on sender and receiver addresses a...

There are many ways to process graph data in deep learning, among which Graph Neural Network(GNN) is an effective and popular deep learning model. However, GNN also has some problems. For example, after multiple layers of neural networks, the features between nodes will become more and more similar, so that the model identifies two completely diffe...

Distributed graph representation learning refers to the process of learning graph data representation in a distributed computing environment. In the process of distributed graph representation learning, nodes need to exchange data frequently, making data transmission crucial in this context. The content of data transmission, including plaintext dat...

Patient representation learning aims to encode meaningful information about the patient’s Electronic Health Records (EHR) in the form of a mathematical representation. Recent advances in deep learning have empowered Patient representation learning methods with greater representational power, allowing the learned representations to significantly imp...

Major depressive disorder (MDD) is the most common psychological disease. To improve the recognition accuracy of MDD, more and more machine learning methods have been proposed to mining EEG features, i.e. typical brain functional patterns and recognition methods that are closely related to depression using resting EEG signals. Most existing methods...

Digital music has various characteristics, such as melody, rhythm, timbre, and harmony. According to these characteristics, music can be classified using artificial intelligence (AI). Music can reduce cognitive dissonance and improve memory in humans; however, occasionally, dissonant music can cause negative effects, such as aggravating depression....

The evaluation and prediction of credit risk have always been a research hotspot to ensure the healthy and orderly development of the credit market. Most researchers use deep learning to predict credit risk. However, when training data are too small, deep learning models often lead to overfitting. Although we have a large amount of available traini...

Unsupervised representation learning approaches aim to learn discriminative feature representations from unlabeled data, without the requirement of annotating every sample. Enabling unsupervised representation learning is extremely crucial for time series data, due to its unique annotation bottleneck caused by its complex characteristics and lack o...

Learning semantic-rich representations from raw unlabeled time series data is critical for downstream tasks such as classification and forecasting. Contrastive learning has recently shown its promising representation learning capability in the absence of expert annotations. However, existing contrastive approaches generally treat each instance inde...

Automatic medical report generation is an essential task in applying artificial intelligence to the medical domain, which can lighten the workloads of doctors and promote clinical automation. The state-of-the-art approaches employ Transformer-based encoder-decoder architectures to generate reports for medical images. However, they do not fully expl...

Insulators are essential and numerous components in power transmission lines, but they are also prone to faults. Therefore, it is crucial to detect faults in insulators. Although existing fault detection methods for insulators in power transmission lines have been improved to some extent by continuously modifying their internal structures, traditio...

Compared with cloud computing, edge–cloud collaboration can avoid long transmitting delay to cloud since tasks are close to edge, which makes edge–cloud collaboration suitable for delay-sensitive applications. However, the complex environment of edge–cloud poses new challenge to task scheduling. A Collaborative Scheduling strategy based on task Adm...

Domain adaptation (DA) aims to transfer knowledge from one source domain to another different but related target domain. The mainstream approach embeds adversarial learning into deep neural networks (DNNs) to either learn domain-invariant features to reduce the domain discrepancy or generate data to fill in the domain gap. However, these adversaria...

Federated Learning (FL), a novel distributed machine learning framework, made it possible to model collaboratively without risking participants’ privacy. All components of FL, including devices, networks, data, and models, are heterogeneous because of the dispersed feature. These heterogeneity issues impeded FL’s performance. HFL (Heterogeneous Fed...

With the accumulation of check-in data from location-based services, next Point-of-Interest (POI) recommendations are gaining increasing attention. It is well known that the spatio-temporal contextual information of user check-in behavior plays a crucial role in handling vital and inherent challenges in next POI recommendation, including capture of...

Federated domain adaptation (FDA) aims to collaboratively transfer knowledge from source clients (domains) to the related but different target client, without communicating the local data of any client. Moreover, the source clients have different data distributions, leading to extremely challenging in knowledge transfer. Despite the recent progress...

Automatic medical image report generation has attracted extensive research interest in medical data mining, which effectively alleviates doctors’ workload and improves report standardization. The mainstream approaches adopt the Transformer-based Encoder-Decoder architecture to align the visual and linguistic features. However, they rarely consider...

Knowledge graph completion is to infer missing/new entities or relations in knowledge graphs. The long-tail distribution of relations leads to the few-shot knowledge graph completion problem. Existing solutions do not thoroughly solve this problem, with the few training samples still deteriorating knowledge graph completion performance. In this pap...

Disease prediction as an important component of medical assistant diagnostic systems has received much attention from researchers. Many studies attempt to extract disease-related features from electronic health records (EHRs). However, the heterogeneity of entities (e.g., diseases, symptoms, medications, and other treatment items) in EHRs is usuall...

Delivery Time Estimation (DTE) is a crucial component of the e-commerce supply chain that predicts delivery time based on merchant information, sending address, receiving address, and payment time. Accurate DTE can boost platform revenue and reduce customer complaints and refunds. However, the imbalanced nature of industrial data impedes previous m...

Delivery Time Estimation (DTE) is a crucial component of the e-commerce supply chain that predicts delivery time based on merchant information, sending address, receiving address, and payment time. Accurate DTE can boost platform revenue and reduce customer complaints and refunds. However, the imbalanced nature of industrial data impedes previous m...

Clinical Phenotyping is a fundamental task in clinical services, which assessments whether a patient suffers a medical condition of interest. Existing works focus on learning better patients’ representations. Recently, multi-task learning has been proposed to transfer knowledge from different tasks and achieved promising performance. However, the e...

Crowdsourcing, in which human intelligence and productivity is dynamically mobilized to tackle tasks too complex for automation alone to handle, has grown to be an important research topic and inspired new businesses (e.g., Uber, Airbnb). Over the years, crowdsourcing has morphed from providing a platform where workers and tasks can be matched up m...

Federated learning has recently been applied to recommendation systems to protect user privacy. In federated learning settings, recommendation systems can train recommendation models only collecting the intermediate parameters instead of the real user data, which greatly enhances the user privacy. Beside, federated recommendation systems enable to...

Learning semantic-rich representations from raw unlabeled time series data is critical for downstream tasks such as classification and forecasting. Contrastive learning has recently shown its promising representation learning capability in the absence of expert annotations. However, existing contrastive approaches generally treat each instance inde...

Mobile crowdsourcing (MC) which has been developed rapidly in recent years is playing an increasingly indispensable role in people’s daily lives such as taxi-hailing, food delivery and other services. The geographic equilibrium of service supply and demand is crucial so that the MC system could guarantee more promising matches in a more regionally...

Numerous electronic health records (EHRs) offer valuable opportunities for understanding patients’ health status at different stages, namely health progression. Extracting the health progression patterns allows researchers to perform accurate predictive analysis of patient outcomes. However, most existing works on this task suffer from the followin...

The sequential recommendation systems capture users' dynamic behavior patterns to predict their next interaction behaviors. Most existing sequential recommendation methods only exploit the local context information of an individual interaction sequence and learn model parameters solely based on the item prediction loss. Thus, they usually fail to l...

Knowledge graph embedding, which aims at learning the low-dimensional representations of entities and relationships, has attracted considerable research efforts recently. However, most knowledge graph embedding methods focus on the structural relationships in fixed triples while ignoring the temporal information. Currently, existing time-aware grap...

The sequential recommendation systems capture users' dynamic behavior patterns to predict their next interaction behaviors. Most existing sequential recommendation methods only exploit the local context information of an individual interaction sequence and learn model parameters solely based on the item prediction loss. Thus, they usually fail to l...

Unsupervised domain adaptation aims to transfer knowledge from a labeled source domain to a related but unlabeled target domain. Most existing approaches either adversarially reduce the domain shift or use pseudo-labels to provide category information during adaptation. However, an adversarial training method may sacrifice the discriminability of t...

Writing medical image reports is an inefficient and time-consuming task for doctors. Automatically generating medical reports is an essential task of medical data mining, which can alleviate the workload of doctors and improve the standardization of reports. However, the existing methods mainly adopt the CNN-RNN structure to align image features wi...

The rapid growth in the use of electronic health records (EHR) offers promises for predicting patient outcomes. Previous works on this task focus on exploiting temporal patterns from sequential EHR data. Nevertheless, such approaches model patients independently, missing out on the similarities between patients, which are crucial for patients’ heal...

Explainable recommendation systems not only provide users with recommended results but also explain why they are recommended. Most existing explainable recommendation methods leverage sentiment analysis to help users understand reasons for recommendation results. They either convert particular preferences into sentiment scores or simply introduce t...

The affordance theory provides a biology-inspired approach to enable a robot to act, think and develop like human beings. Based on existing affordance relationships, a robot understands its environment and task in terms of potential actions that it can execute. Deep learning makes it possible for a robot to perceive the environment in an efficient...

In recent years, neural networks have been widely used in natural language processing, especially in sentence similarity modeling. Most of the previous studies focused on the current sentence, ignoring the commonsense knowledge related to the current sentence in the task of sentence similarity modeling. Commonsense knowledge can be remarkably usefu...

The target of product attributes prediction is to complete the characteristics set for defining a particular product. Most of the existing methods treat the product attributes prediction as a Named-Entity Recognition (NER) problem from the products’ affiliated data, such as product title and introduction. However, in a large number of industrial ap...

Vertical Federated Learning (VFL) is a private-preserving distributed machine learning paradigm that collaboratively trains machine learning models with participants whose local data overlap largely in the sample space, but not so in the feature space. Existing VFL methods are mainly based on synchronous computation and homomorphic encryption (HE)....

On the increase of major depressive disorders (MDD), many researchers paid attention to their recognition and treatment. Existing MDD recognition algorithms always use a single time-frequency domain method method, but the single time-frequency domain method is too simple and is not conducive to simulating the complex link relationship between brain...

Graph neural networks (GNN) have recently been applied to exploit knowledge graph (KG) for recommendation. Existing GNN-based methods explicitly model the dependency between an entity and its local graph context in KG (i.e., the set of its first-order neighbors), but may not be effective in capturing its non-local graph context (i.e., the set of mo...

Power iteration has been applied to compute the eigenvectors of the similarity matrix in spectral clustering tasks. However, these power iteration based clustering methods usually suffer from the following two problems: (1) the power iteration usually converges very slowly; (2) the singular value decomposition method adopted to obtain the eigenvect...

Knowledge graph embedding, which aims to learn the low-dimensional representations of entities and relationships, has attracted considerable research efforts recently. However, most knowledge graph embedding methods focus on the structural relationships in fixed triples while ignoring the temporal information. Currently, existing time-aware graph e...

Transfer learning (TL) is a machine learning paradigm designed for the problem where the training and test data are from different domains. Existing TL approaches mostly assume that training data from the source domain are collected from multiple views or devices. However, in practical applications, a sample in a target domain often only correspond...

Graph neural networks (GNN) have recently been applied to exploit knowledge graph (KG) for recommendation. Existing GNN-based methods explicitly model the dependency between an entity and its local graph context in KG (i.e., the set of its first-order neighbors), but may not be effective in capturing its non-local graph context (i.e., the set of mo...

Deep learning has proven to be effective in learning scenarios with massive training data. However, in many real applications (i.e., abnormality detection), there is a lack of sufficient data to achieve a good deep learning model. Considering the fact that collecting massive labeled training data for a new task is often expensive and time-consuming...

Multi-label learning deals with problems in which each instance is associated with a set of labels. Most multi-label learning algorithms ignore the potential distribution differences between the training domain and test domain in the instance space and label space, as well as the intrinsic geometric information of the label space. These restrictive...

Multi-Instance learning (MIL) aims to predict labels of unlabeled bags by training a model with labeled bags. The usual assumption of existing MIL methods is that the underlying distribution of training data is the same as that of the testing data. However, this assumption may not be valid in practice, especially when training data from a source do...

Background and Purpose: A majority studies on diagnosis of Alzheimer’s Disease (AD) are based on an assumption: the training and testing data are drawn from the same distribution. However, in the diagnosis of AD and mild cognitive impairment (MCI), this identical-distribution assumption may not hold. To solve this problem, we utilize the transfer l...

Transfer learning techniques have been broadly applied in applications where labeled data in a target domain are difficult to obtain while a lot of labeled data are available in related source domains. In practice, there can be multiple source domains that are related to the target domain, and how to combine them is still an open problem. In this p...

Transfer learning has been proven to be effective for the problems where training data from a source domain and test data from a target domain are drawn from different distributions. To reduce the distribution divergence between the source domain and the target domain, many previous studies have been focused on designing and optimizing objective fu...

Multi-Instance (MI) learning has been proven to be effective for the genome-wide protein function prediction problems where each training example is associated with multiple instances. Many studies in this literature attempted to find an appropriate Multi-Instance Learning (MIL) method for genome-wide protein function prediction under a usual assum...

Multi-instance multi-label (MIML) learning has been proven to be effective for the genome-wide protein function prediction problems where each training example is associated with not only multiple instances but also multiple class labels. To find an appropriate MIML learning method for genome-wide protein function prediction, many studies in the li...

Query-URL relevance, measuring the relevance of each retrieved URL with respect to a given query, is one of the fundamental criteria to evaluate the performance of commercial search engines. The traditional way to collect reliable and accurate query-URL relevance requires multiple annotators to provide their individual judgments based on their subj...