Search

Temporal graphs exhibit dynamic interactions between nodes over continuous time, whose topologies evolve with time elapsing. The whole temporal neighborhood of nodes reveals the varying preferences of nodes. However, previous works usually generate dynamic representation with limited neighbors for simplicity, which results in both inferior performance and high latency of online inference. Therefore, in this paper, we propose a novel method of temporal graph convolution with the whole neighborhood, namely Temporal Aggregation and Propagation Graph Neural Networks (TAP-GNN). Specifically, we first analyze the computational complexity of the dynamic representation problem by unfolding the temporal graph in a message-passing paradigm. The expensive complexity motivates us to design the AP (aggregation and propagation) block, which significantly reduces the repeated computation of historical neighbors. The final TAP-GNN supports online inference in the graph stream scenario, which incorporates the temporal information into node embeddings with a temporal activation function and a projection layer besides several AP blocks. Experimental results on various real-life temporal networks show that our proposed TAP-GNN outperforms existing temporal graph methods by a large margin in terms of both predictive performance and online inference latency.

Graph disentangling is a new promising direction that can help us to discover the latent patterns in the data and understand the behaviors of a graph learning model. Despite the many efforts in disentangling representation learning, few works focus on disentangling the latent factors behind a graph. Most current foci are mainly on studying node-level semantics in the graphs. Compared with node-level, the structure-level view can provide a new interpretable and in-depth insight into graph data. The study of structure-level relations enables us to reveal the high-order structural semantics in the data. To explore the complex high-order structural semantics in the data, we propose the High-order Structural Semantic Disentangled Neural Network (HSDN) to model the graph structure units and disentangle structural semantics. It's the first attempt to hypergraph disentangled networks. Unlike prior methods that disentangle factor graphs based on pair-wise relations only, we introduce hyperedges on pair-wise graphs to model structure units and disentangle the complex high-order structural semantics between different structures. Extensive experiments demonstrate that HSDN achieves state-of-the-art performances in terms of both disentangling and downstream tasks.

We explored the application of TFE3 immunostaining and TFE3 mRNA expression in the differential diagnosis and prognostication of adrenal cortical tumors and distinction of the latter from clear cell renal cell carcinoma (ccRCC) which show significant morphologic overlap. TFE3 immunostaining was performed on a large cohort of samples including 40 adrenal cortex tissues, 95 adrenocortical adenoma (ACA), 11 adrenocortical carcinoma (ACC), 53 ccRCC, and 18 pheochromocytomas. TFE3 was compared with other immunomarkers melan-A, inhibin-α, synaptophysin, chromogranin A, CAIX and CD10. One hundred percent normal adrenal cortices and 94% ACA were strongly and diffusely stained for TFE3 while no ACC showed diffuse staining. TFE3 is thus useful in distinguishing ACA from ACC. TFE3 is also useful in separating ACC from ccRCC as 64% ACC showed partial, while only 7% of ccRCC showed partial TFE3 staining. Only 1 pheochromocytoma showed focal weak TFE3 staining. Results also demonstrated superiority of TFE3 over other commonly used immunomarkers. TFE3 gene rearrangement testing by fluorescence in situ hybridization showed no rearrangement in 6 TFE3 positive adrenal tumors. TFE3 mRNA were analyzed by the Cancer Genome Atlas database and we found TFE3 mRNA expression correlated with overall patient survival in ACC. Our study showed usefulness of TFE3 in distinguishing ACA from ACC, and ACC from ccRCC. TFE3 is superior over other commonly used immunomarkers for adrenal tumors. In addition, decreased TFE3 immunoexpression and TFE3 mRNA expression may carry poor prognostic implication in adrenal tumors.

Although deep learning has demonstrated astonishing performance in many applications, there are still concerns about its dependability. One desirable property of deep learning applications with societal impact is fairness (i.e., non-discrimination). Unfortunately, discrimination might be intrinsically embedded into the models due to the discrimination in the training data. As a countermeasure, fairness testing systemically identifies discriminatory samples, which can be used to retrain the model and improve the model’s fairness. Existing fairness testing approaches however have two major limitations. First, they only work well on traditional machine learning models and have poor performance (e.g., effectiveness and efficiency) on deep learning models. Second, they only work on simple structured (e.g., tabular) data and are not applicable for domains such as text. In this work, we bridge the gap by proposing a scalable and effective approach for systematically searching for discriminatory samples while extending existing fairness testing approaches to address a more challenging domain, i.e., text classification. Compared with state-of-the-art methods, our approach only employs lightweight procedures like gradient computation and clustering, which is significantly more scalable and effective. Experimental results show that on average, our approach explores the search space much more effectively (9.62 and 2.38 times more than the state-of-the-art methods respectively on tabular and text datasets) and generates much more discriminatory samples (24.95 and 2.68 times) within a same reasonable time. Moreover, the retrained models reduce discrimination by 57.2 and 60.2 percent respectively on average.

To characterize the clinicopathologic features of pulmonary sclerosing pneumocytoma (PSP) and compare these features between the tumors with and without metastasis, 68 cases of PSP (1/68 [1.47%] with metastasis) diagnosed from 2009-2022 in our hospital and 15 previously reported metastasizing cases were studied. There were 54 female patients and 14 male patients, with age ranging from 17 to 72 years and tumor size ranging from 0.1 to 5.5 cm (mean, 1.75 cm). In all, 85.4% of the cases presented with ≥2 patterns, including papillary, sclerotic, solid, and hemorrhagic. Thyroid transcription factor 1, epithelial membrane antigen, CKpan, and CK7 were expressed in surface cells in 100% of the cases and napsin A was expressed in 90% of the cases. Stromal cell expression of these markers occurred in 100%, 93.9%, 13.5%, 13.8%, and 0% of the cases, respectively. Of the 16 PSP cases with metastasis, 8 were female patients and 7 were male patients, with age ranging from 14 to 73 years. The tumor size ranged from 2.5 to 12 cm (mean, 4.85 cm). Forty-five of the cases were negative for BRAF V600E immunostaining and 6 were focally weak positive, in which fluorescent PCR tests showed no detectable mutations. There were significant differences in gender, age, and tumor size between PSP cases with and without metastasis. No BRAFV600E mutation was found in patients with PSP. AKT1p.E17K mutations were detected in both the primary lung tumor and the lymph node metastatic tumor of our PSP case with lymph node metastasis. In conclusion, PSP is an uncommon pulmonary neoplasm with significant female predilection and has distinct morphologic and immunohistochemical characteristics. The BRAFV600Emutation was not detectable in patients with PSP and thus may not involve in its tumorigenesis. Most PSP tumors are benign, with a minority exhibiting potential for metastasis and malignant behavior.

In this article, a usage model is proposed to simulate users’ behaviors realistically in load testing of web applications, and another relevant workload model is proposed to help generate realistic loads for load testing. It also demonstrates an eclipse-based load testing tool ‘Load Testing Automation Framework’ which is based on these two models and can perform load testing of web applications easily and automatically. Furthermore, these models and tools were successfully applied into a representative web-based system from a big corporation.

Human beings keep exploring the physical space using information means. Only recently, with the rapid development of information technologies and the increasing accumulation of data, human beings can learn more about the unknown world with data-driven methods. Given data timeliness, there is a growing awareness of the importance of real-time data. There are two categories of technologies accounting for data processing: batching big data and streaming processing, which have not been integrated well. Thus, we propose an innovative incremental processing technology named after Stream Cube to process both big data and stream data. Also, we implement a real-time intelligent data processing system, which is based on real-time acquisition, real-time processing, real-time analysis, and real-time decision-making. The real-time intelligent data processing technology system is equipped with a batching big data platform, data analysis tools, and machine learning models. Based on our applications and analysis, the real-time intelligent data processing system is a crucial solution to the problems of the national society and economy.