Dissolved Gas Analysis (DGA) is a widely used diagnostic technique specifically designed to identify potential defects or irregularities in power transformers. Despite recent advancements, the fine-grained correlation modeling between samples is still insufficient. This limitation results in less contextualized feature representations, ultimately impacting the accuracy of diagnosis. To this end, we propose an effective graph deep learning model called DGRCL (Dual-channel Graph Ranking Contrastive Learning) for DGA. In our model, we construct two types of graphs: similar$K$-Nearest Neighbors (KNN) ($G_{S}$) and dissimilarKNN ($G_{\text{DS}}$), where nodes represent individual records of dissolved gases and the edges, namely similar-homophilic and dissimilar-heterophilic, capture the relationships between nodes. To better generate node representations, we design a dual-channel graph encoding component, with each channel playing a crucial role in capturing the corresponding edge type view. Furthermore, in order to better distinguish the feature differences among various fault categories, we employ ranking contrastive loss to refine the node representations. Experimental results demonstrate that our model achieves high accuracy and robustness in predicting fault types.

Query-aware synthetic data generation is an essential and highly challenging task, important for database management system (DBMS) testing, database application testing and application-driven benchmarking. Prior studies on query-aware data generation suffer common problems of limited parallelization, poor scalability, and excessive memory consumption, making these systems unsatisfactory to terabyte scale data generation. In order to fill the gap between the existing data generation techniques and the emerging demands of enormous query-aware test databases, we design and implement a new data generator, called Touchstone. Touchstone adopts the random sampling algorithm instantiating query parameters and the new data generation schema generating the test database, to achieve fully parallel data generation, linear scalability and austere memory consumption. It has full support of outer joins as well as non-equi-joins for application-oriented data generation. Our experimental results show that Touchstone consistently outperforms the state-of-the-art solution on TPC-H workload by a 1000 speedup without sacrificing simulation fidelity.

Generating synthetic workloads is essential and critical to performance evaluation of database systems. When evaluating database performance for a specific application, the similarity between synthetic workloads and real application workloads determines the credibility of the evaluation results. However, it meets a great challenge to catch workload characteristics with respect to a target application considering the complexity of transaction executions. To address this problem, we propose a workload duplicator (Lauca) that can generate synthetic workloads with highly similar performance metrics compared to the real workloads of a specific application. By carefully studying the application-oriented workload generation problem, we present Transaction Logic and Data Access Distribution to characterize workloads of online transaction processing (OLTP) applications, and propose novel generation algorithms to guarantee the high fidelity of synthetic workloads. To the best of our knowledge, Lauca is the first application-oriented transactional workload generator. We conduct extensive experiments based on TPCC, SmallBank and YCSB on both centralized and distributed databases. The experimental results show that Lauca consistently generates high quality synthetic workloads.

Food retrieval is an important task to perform analysis of food-related information, where we are interested in retrieving relevant information about the queried food item such as ingredients, cooking instructions, etc. In this paper, we investigate cross-modal retrieval between food images and cooking recipes. The goal is to learn an embedding of images and recipes in a common feature space, such that the corresponding image-recipe embeddings lie close to one another. Two major challenges in addressing this problem are 1) large intra-variance and small inter-variance across cross-modal food data; and 2) difficulties in obtaining discriminative recipe representations. To address these two problems, we propose Semantic-Consistent and Attention-based Networks (SCAN), which regularize the embeddings of the two modalities through aligning output semantic probabilities. Besides, we exploit a self-attention mechanism to improve the embedding of recipes. We evaluate the performance of the proposed method on the large-scale Recipe1M dataset, and show that we can outperform several state-of-the-art cross-modal retrieval strategies for food images and cooking recipes by a significant margin.

Face anti-spoofing is becoming increasingly indispensable for face recognition systems, which are vulnerable to various spoofing attacks performed using fake photos and videos. In this paper, a novel “LDN-TOP representation followed by ProCRC classification” pipeline for face anti-spoofing is proposed. We use local directional number pattern (LDN) with the derivative-Gaussian mask to capture detailed appearance information resisting illumination variations and noises, which can influence the texture pattern distribution. To further capture motion information, we extend LDN to a spatial-temporal variant named local directional number pattern from three orthogonal planes (LDN- TOP). The multi-scale LDN- TOP capturing complete information is extracted from color images to generate the feature vector with powerful representation capacity. Finally, the feature vector is fed into the probabilistic collaborative representation based classifier (ProCRC) for face anti-spoofing. Our method is evaluated on three challenging public datasets, namely CASIA FASD, Replay-Attack database, and UVAD database using sequence-based evaluation protocol. The experimental results show that our method can achieve promising performance with 0.37% EER on CASIA and 5.73% HTER on UVAD. The performance on Replay-Attack database is also competitive.