Public-private graph, where a public network is visible to everyone and every user is also associated with its own small private graph accessed by itself only, widely exists in real-world applications of social networks and financial networks. Most existing work on community search, finding a query-dependent community containing a given query, only studies on a public graph, neglecting the privacy issues in public-private networks. However, considering both the public and private attributes of users enables community search to be more accurate, comprehensive, and personalized to discover hidden patterns. In this paper, we study a novel problem of attributed community search in public-private graphs (ACS-PP), aiming to find a connected k-core community that shares the most keywords with the query node. This problem uncovers structurally cohesive communities, such as interest-based user groups or core teams in collaborative networks. To optimize search efficiency, we propose an integrated scheme of constructing a public global graph index and a private personalized graph index. For the private index, we developed a compact structure of the PP-FP-tree index. The PP-FP-tree is constructed based on the public and private neighbors of the query node in the public-private graph, serving as an efficient index to mine frequent node sets that share the most common attributes with the query node. Extensive experiments on real public-private graph datasets validate both the efficiency and quality of our proposed PP-FP search algorithm against existing competitors. The case study on public-private collaboration networks provides insights into the discovery of public-private communities.

Explorative flow visualization allows domain experts to analyze complex flow structures by interactively investigating flow patterns. However, traditional visual interfaces often rely on specialized graphical representations and interactions, which require additional effort to learn and use. Natural language interaction offers a more intuitive alternative, but teaching machines to recognize diverse scientific concepts and extract corresponding structures from flow data poses a significant challenge. In this paper, we introduce an automated framework that aligns flow pattern representations with the semantic space of large language models (LLMs), eliminating the need for manual labeling. Our approach encodes streamline segments using a denoising autoencoder and maps the generated flow pattern representations to LLM embeddings via a projector layer. This alignment empowers semantic matching between textual embeddings and flow representations through an attention mechanism, enabling the extraction of corresponding flow patterns based on textual descriptions. To enhance accessibility, we develop an interactive interface that allows users to query and visualize flow structures using natural language. Through case studies, we demonstrate the effectiveness of our framework in enabling intuitive and intelligent flow exploration.

Identifying biomarkers from human brain networks is critical in early detection of neurological disorders and understanding disease mechanisms. Existing visual analytics approaches show a remarkable ability to assist experts in discovering and validating biomarkers through exploration. However, these approaches often focus only on the diffusion features of fiber bundles and evaluate individual bundles and regions separately. This may hinder their ability to accurately represent the complex brain network for investigation from various perspectives. In this paper, we present NeuroLens, a visual analytics system that integrates comprehensive information across multiple levels, including fiber bundles, local regions and entire brain networks. Specifically, to model the bundles more precisely, we enhance the features by incorporating the joint distribution of geometric features. The bundle information is further aggregated to form representations at the region and the brain level using an attention-based graph neural network. The region-level representation describes complex structures involving multiple bundles, and the brain-level representation enables comparisons between subjects and groups. The NeuroLens interface enables comparative exploration of this multi-level and multi-faceted information. To verify the findings during exploration, NeuroLens leverages the large language model to query related information from existing literature. We collaborate with domain experts to examine the effectiveness of NeuroLens. Their exploration, findings, and feedback are discussed.

Deep joint source-channel coding (DeepJSCC) has emerged as a novel technology in semantic communication, coinciding with the increasing demand for the edge devices in the Internet of Things (IoT). Consequently, the deployment of DeepJSCC on edge devices has become a crucial research direction. However, DeepJSCC faces challenges related to channel fading. Moreover, implementing DeepJSCC on the edge devices poses challenges due to the constrained computational resources as well as the compatibility issue between DeepJSCC and digital systems. In this article, we devote to engineering the DeepJSCC system deployed on the edge devices. First, we propose a method named DeepJSCC with Ensemble learning (DeepJSCC-ES) to resist the channel fading. Then, we present a pruning algorithm called the DeepJSCC signal-to-noise ratio (SNR)-adaptive pruning method (DJSAP) to make the DeepJSCC network lightweight, reducing the computational demands on the edge nodes. Further, we propose a method called the simulated fixed-point quantization training based on soft quantization function (SFPQSQ) to tackle the compatibility issue between DeepJSCC and digital systems. Finally, we deploy the whole DeepJSCC system on the edge devices and conduct experiments to test the DeepJSCC system. The results of simulations show that the proposed DeepJSCC-ES system outperforms the baseline DeepJSCC, particularly excelling in low SNR conditions. Furthermore, the parameter size of the pruned model using DJSAP is compressed by 93.37% while the average structural similarity index metric (SSIM) decreases only by 0.92% compared with the baseline DeepJSCC. Additionally, the SFPQSQ works better than the ordinary quantization methods in tackling the compatibility issue between DeepJSCC and digital systems. The experiment results also show that our proposed system can serve as a feasible solution for practical deployment on the edge devices.

Ordering has been extensively studied in many visualization applications, such as axis and matrix reordering, for the simple reason that the order will greatly impact the perceived pattern of data. Many quality metrics concerning data pattern, perception, and aesthetics are proposed, and respective optimization algorithms are developed. However, the optimization problems related to ordering are often difficult to solve (e.g., TSP is NP-complete), and developing specialized optimization algorithms is costly. In this paper, we propose Versatile Ordering Network (VON), which automatically learns the strategy to order given a quality metric. VON uses the quality metric to evaluate its solutions, and leverages reinforcement learning with a greedy rollout baseline to improve itself. This keeps the metric transparent and allows VON to optimize over different metrics. Additionally, VON uses the attention mechanism to collect information across scales and reposition the data points with respect to the current context. This allows VONs to deal with data points following different distributions. We examine the effectiveness of VON under different usage scenarios and metrics. The results demonstrate that VON can produce comparable results to specialized solvers.

Flow visualization is an essential tool for domain experts to understand and analyze flow fields intuitively. In the past decades, various interactive techniques were developed to customize flow visualization for exploration. However, these techniques usually use specifically designed graphical interfaces, requiring considerable learning and usage effort. Recently, FlowNL Huang et al., (2023) introduces a natural language interface to reduce the effort, but it still struggles with natural language ambiguities due to the lack of domain knowledge and provides limited ability to understand the context in dialogues. To address these issues, we propose an explorative flow visualization powered by a large language model that interacts with users. Our approach leverages an extensive dataset of flow-related queries to train the model, enhancing its ability to interpret a wide range of natural language expressions and maintain context over multi-turn interactions. Additionally, we introduce an advanced dialogue management system that supports interactive continuous communication between users and the system. Our empirical evaluations demonstrate significant improvements in user engagement and accuracy of flow structure extraction. These enhancements are crucial for expanding the applicability of flow visualization systems in real-world scenarios, where effective and intuitive user interfaces are paramount.

Task-oriented semantic communications introduce a novel paradigm specifically designed to enhance task-specific performance. However, this paradigm may face limitations as it requires frequent updating with task changes or necessitates storing multiple distinct models for various tasks. To address these challenges, we propose a task-adaptive semantic communication system with feedback (TASC-f), which utilizes a single model capable of adapting to variable tasks. In particular, we formulate a conditional rate-distortion optimization problem, where task-specific prompts serve as dynamic side information to guide coding strategies and enhance task performance. Inspired by visual prompt tuning, we present a learnable semantic prompt model (SPM) coupled with a dynamic parameters network, aimed at effectively extracting task-specific features. A feedback mechanism is also integrated to capture real-time task information, facilitating timely adjustments of the coding policy. In our experiments, we employ the TASC-f system to evaluate its effectiveness across three AI tasks within two distinct scenarios: tasks that are newly introduced and those previously encountered during the training phase. Simulation results show that our proposed TASC-f surpasses all data-oriented communication schemes in both scenarios and achieves performance comparable to single-task-oriented semantic systems with reduced communication overhead and fewer model parameters.

Abstract It is a critical quality assurance activity to effectively detect the root cause of faults causing the software crashes (i.e. crashing faults). Previous studies extracted features to characterise crash instances and built models to identify whether the residences of crashing faults locate inside the stack traces. These models all belong to supervised learning methods which require labelled crash data to be involved. In this study, the introduction of an unsupervised model, called T ransfer S pectral C lustering ( TSC ), for the task of crashing fault residence identification under the unlabelled data scenario is proposed. Unlike traditional unsupervised methods which are applied to individual project data, TSC transfers the knowledge of auxiliary unlabelled data from the source project to assist the clustering task on the unlabelled data from the target project. TSC is an unsupervised transfer learning method, and simultaneously considers the data manifold information of the individual project and feature manifold information across projects to facilitate the clustering effect. Extensive experiments are conducted on a benchmark dataset containing seven software projects. Five indicators were chosen for performance evaluation. The results show that TSC achieves better performance than four clustering based unsupervised methods, and competitive performance compared with eight supervised cross‐project methods.