Mutual funds are one of the most important and popular investment ways for ordinary investors to maintain and increase the value of their assets. However, it is challenging for ordinary investors to select optimal mutual funds from thousands of fund choices managed by different managers. Various investors often have different personal investment preferences and it is difficult to characterize their preferences quickly. Also, mutual fund performance relies on various factors (e.g., the economic market and the management of fund managers), and most of these factors are dynamically changing, making it difficult to efficiently compare different mutual funds in detail. To address these challenges, we propose FundSelector, an interactive multi-view visual analytics system that quantifies user preferences to rank mutual funds and allows ordinary investors to explore mutual fund performance in terms of multiple factors and scales. Two novel visual designs are proposed to enable detailed comparisons of mutual funds. Rank-informed bipartite contribution bar chart provides interpretable fund ranking results by explicitly showing both positive and negative factors. Elastic trend chart allows investors to analyze and compare the temporal evolution of the mutual funds’ performances in a customizable way. We evaluated FundSelector through two case studies and interviews with eight ordinary investors. The results highlight its effectiveness and utility.

CTR prediction is critical to industrial recommender systems. Recently, with the growth of business domains in enterprises, much attention has been focused on the multi-domain CTR recommendation. Numerous models have been proposed that attempt to use a unified model to serve multiple domains. Although much progress has been made, we argue that they ignore the importance of feature interactions and user behaviors when modeling cross-domain relations, which is a coarse-grained utilizing of domain information. To solve this problem, we propose Domain Facilitated Feature Modeling (DFFM) for CTR prediction. It incorporates domain-related information into the parameters of the feature interaction and user behavior modules, allowing for domain-specific learning of these two aspects. Extensive experiments are conducted on two public datasets and one industrial dataset to demonstrate the effectiveness of DFFM. We deploy the DFFM model in Huawei advertising platform and gain a 4.13% improvement of revenue on a two week online A/B test. Currently DFFM model has been used as the main traffic model, serving for hundreds of millions of people.

Multi-Domain Recommendation (MDR) has gained significant attention in recent years, which leverages data from multiple domains to enhance their performance concurrently. However, current MDR models are confronted with two limitations. Firstly, the majority of these models adopt an approach that explicitly shares parameters between domains, leading to mutual interference among them. Secondly, due to the distribution differences among domains, the utilization of static parameters in existing methods limits their flexibility to adapt to diverse domains. To address these challenges, we propose a novel model HAMUR. Specifically, HAMUR consists of two components: (1). Domain-specific adapter, designed as a pluggable module that can be seamlessly integrated into various existing multi-domain backbone models, and (2). Domain-shared hyper-network, which implicitly captures shared information among domains and dynamically generates the parameters for the adapter. We conduct extensive experiments on two public datasets using various backbone networks. The experimental results validate the effectiveness and scalability of the proposed model.

Sequential recommendation (SRS) has become the technical foundation in many applications recently, which aims to recommend the next item based on the user's historical interactions. However, sequential recommendation often faces the problem of data sparsity, which widely exists in recommender systems. Besides, most users only interact with a few items, but existing SRS models often underperform these users. Such a problem, named the long-tail user problem, is still to be resolved. Data augmentation is a distinct way to alleviate these two problems, but they often need fabricated training strategies or are hindered by poor-quality generated interactions. To address these problems, we propose a Diffusion Augmentation for Sequential Recommendation (DiffuASR) for a higher quality generation. The augmented dataset by DiffuASR can be used to train the sequential recommendation models directly, free from complex training procedures. To make the best of the generation ability of the diffusion model, we first propose a diffusion-based pseudo sequence generation framework to fill the gap between image and sequence generation. Then, a sequential U-Net is designed to adapt the diffusion noise prediction model U-Net to the discrete sequence generation task. At last, we develop two guide strategies to assimilate the preference between generated and origin sequences. To validate the proposed DiffuASR, we conduct extensive experiments on three real-world datasets with three sequential recommendation models. The experimental results illustrate the effectiveness of DiffuASR. As far as we know, DiffuASR is one pioneer that introduce the diffusion model to the recommendation.The implementation code is available online.

This paper reports an empirical study of the effectiveness of three different authentication methods for Internet of Things (IoT) networks. They are simulated under three attack strategies with three different structural topologies. The survival size and survival link ratio (SLR) are examined to provide some references and insight for designing IoT networks under various conditions and applications.

A non-fungible token (NFT) is a data unit stored on the blockchain. Nowadays, more and more investors and collectors (NFT traders), who participate in transactions of NFTs, have an urgent need to assess the performance of NFTs. However, there are two challenges for NFT traders when analyzing the performance of NFT. First, the current rarity models have flaws and are sometimes not convincing. In addition, NFT performance is dependent on multiple factors, such as images (high-dimensional data), history transactions (network), and market evolution (time series). It is difficult to take comprehensive consideration and analyze NFT performance efficiently. To address these challenges, we propose NFTVis, a visual analysis system that facilitates assessing individual NFT performance. A new NFT rarity model is proposed to quantify NFTs with images. Four well-coordinated views are designed to represent the various factors affecting the performance of the NFT. Finally, we evaluate the usefulness and effectiveness of our system using two case studies and user studies.

Cross-Domain Recommendation (CDR) is a widely used approach for leveraging information from domains with rich data to assist domains with insufficient data. A key challenge of CDR research is the effective and efficient transfer of helpful information from source domain to target domain. Currently, most existing CDR methods focus on extracting implicit information from the source domain to enhance the target domain. However, the hidden structure of the extracted implicit information is highly dependent on the specific CDR model, and is therefore not easily reusable or transferable. Additionally, the extracted implicit information only appears within the intermediate substructure of specific CDRs during training and is thus not easily retained for more use. In light of these challenges, this paper proposes AutoTransfer, with an Instance Transfer Policy Network, to selectively transfers instances from source domain to target domain for improved recommendations. Specifically, AutoTransfer acts as an agent that adaptively selects a subset of informative and transferable instances from the source domain. Notably, the selected subset possesses extraordinary re-utilization property that can be saved for improving model training of various future RS models in target domain. Experimental results on two public CDR benchmark datasets demonstrate that the proposed method outperforms state-of-the-art CDR baselines and classic Single-Domain Recommendation (SDR) approaches. The implementation code is available for easy reproduction.

Two-tower models are a prevalent matching framework for recommendation, which have been widely deployed in industrial applications. The success of two-tower matching attributes to its efficiency in retrieval among a large number of items, since the item tower can be precomputed and used for fast Approximate Nearest Neighbor (ANN) search. However, it suffers two main challenges, including limited feature interaction capability and reduced accuracy in online serving. Existing approaches attempt to design novel late interactions instead of dot products, but they still fail to support complex feature interactions or lose retrieval efficiency. To address these challenges, we propose a new matching paradigm named SparCode, which supports not only sophisticated feature interactions but also efficient retrieval. Specifically, SparCode introduces an all-to-all interaction module to model fine-grained query-item interactions. Besides, we design a discrete code-based sparse inverted index jointly trained with the model to achieve effective and efficient model inference. Extensive experiments have been conducted on open benchmark datasets to demonstrate the superiority of our framework. The results show that SparCode significantly improves the accuracy of candidate item matching while retaining the same level of retrieval efficiency with two-tower models.

Automatic visualization generates meaningful visualizations to support data analysis and pattern finding for novice or casual users who are not familiar with visualization design. Current automatic visualization approaches adopt mainly aggregation and filtering to extract patterns from the original data. However, these limited data transformations fail to capture complex patterns such as clusters and correlations. Although recent advances in feature engineering provide the potential for more kinds of automatic data transformations, the auto-generated transformations lack explainability concerning how patterns are connected with the original features. To tackle these challenges, we propose a novel explainable recommendation approach for extended kinds of data transformations in automatic visualization. We summarize the space of feasible data transformations and measures on explainability of transformation operations with a literature review and a pilot study, respectively. A recommendation algorithm is designed to compute optimal transformations, which can reveal specified types of patterns and maintain explainability. We demonstrate the effectiveness of our approach through two cases and a user study.

We present ONCE-3DLanes, a real-world autonomous driving dataset with lane layout annotation in 3D space. Conventional 2D lane detection from a monocular image yields poor performance of following planning and control tasks in autonomous driving due to the case of uneven road. Predicting the 3D lane layout is thus necessary and enables effective and safe driving. However, existing 3D lane detection datasets are either unpublished or synthesized from a simulated environment, severely hampering the development of this field. In this paper, we take steps towards addressing these issues. By exploiting the explicit relationship between point clouds and image pixels, a dataset annotation pipeline is designed to automatically generate high-quality 3D lane locations from 2D lane annotations in 211K road scenes. In addition, we present an extrinsic-free, anchorfree method, called SALAD, regressing the 3D coordinates of lanes in image view without converting the feature map into the bird's-eye view (BEV). To facilitate future research on 3D lane detection, we benchmark the dataset and provide a novel evaluation metric, performing extensive experiments of both existing approaches and our proposed method. The aim of our work is to revive the interest of 3D lane detection in a real-world scenario. We believe our work can lead to the expected and unexpected innovations in both academia and industry.

Neurofibromatosis type 1 (NF1) is an autosomal dominant tumor predisposition syndrome that involves the central and peripheral nervous systems. Accurate detection and segmentation of neurofibromas are essential for assessing tumor burden and longitudinal tumor size changes. Automatic convolutional neural networks (CNNs) are sensitive and vulnerable as tumors' variable anatomical location and heterogeneous appearance on MRI. In this study, wepropose deep interactive networks (DINs) to address the above limitations. User interactions guide the model to recognize complicated tumors and quickly adapt to heterogeneous tumors. We introduce a simple but effective Exponential Distance Transform (ExpDT) that converts user interactions into guide maps regarded as the spatial and appearance prior. Comparing with popular Euclidean and geodesic distances, ExpDT is more robust to various image sizes, which reserves the distribution of interactive inputs. Furthermore, to enhance the tumor-related features, we design a deep interactive module to propagate the guides into deeper layers. We train and evaluate DINs on three MRI data sets from NF1 patients. The experiment results yield significant improvements of 44% and 14% in DSC comparing with automated and other interactive methods, respectively. We also experimentally demonstrate the efficiency of DINs in reducing user burden when comparing with conventional interactive methods.

Code summarization aims to summarize code functionality as high-level nature language descriptions to assist in code comprehension. Recent approaches in this field mainly focus on generating summaries for code with precise identifier names, in which meaningful words can be found indicating code functionality. When faced with lexically confusing code, current approaches are likely to fail since the correlation between code lexical tokens and summaries is scarce. To tackle this problem, we propose a novel summarization framework named VECOS. VECOS introduces an erosion mechanism to conquer the model's reliance on precisely defined lexical information. To facilitate learning the eroded code's functionality, we force the representation of the eroded code to align with the representation of its original counterpart via variational inference. Experimental results show that our approach outperforms the state-of-the-art approaches to generate coherent and reliable summaries for various lexically confusing code.

A large amount of malware families use the domain generation algorithms (DGA) to randomly generate a large amount of domain names. It is a good way to bypass conventional blacklists of domain names, because we cannot predict which of the randomly generated domain names are selected for command and control (C&C) communications. An effective approach for detecting known DGA families is to investigate the malware with reverse engineering to find the adopted generation algorithms. As reverse engineering cannot handle the variants of DGA families, some researches leverage supervised learning to find new variants. However, the explainability of supervised learning is low and cannot find previously unseen DGA families. In this paper, we propose a graph-based semi-supervised learning scheme to track the evolution of known DGA families and find previously unseen DGA families. With a domain relation graph, we can clearly figure out how new variants relate to known DGA domain names, which induces better explainability. We deployed the proposed scheme on real network scenarios and show that the proposed scheme can not only comprehensively and precisely find known DGA families, but also can find new DGA families which have not seen before.

For opportunistic routing in independent duty-cycled wireless sensor networks (WSNs), a sender dynamically determines a relay candidate set depending on the real-time network conditions. Due to independent and varying duty cycle length, the sender may handle different waking-up orders of the potential forwarders when it tries to forward data packets at different time instants. Conventional opportunistic routing protocols overlook the time-varying property of the waking-up order of the candidate nodes. In this paper, we theoretically analyze how to obtain an optimal candidate set for each node in order to minimize the end-to-end latency. Then, considering the realistic scenarios, we propose an opportunistic routing which jointly considers global and localized optimizations. Based on the relatively stable topology and duty-cycle length information, an original candidate set is constructed. Then, by considering the real-time link and duty cycle information in the local context, a further optimization for the original candidate set can be achieved. Simulation results show that our proposed schemes can significantly improve the end-to-end latency compared with the benchmarks.

Network coding and schedule-based flooding trees have been widely deployed to improve the performance of a flooding process in low-duty-cycle wireless sensor networks (WSNs). However, the conventional schemes induce too long flooding delay because there is only one static sink node as a single source. In this paper, we use a mobile sink to travel around the monitoring area. When the mobile sink travels to a position, it serves as a new source. Since multiple sources can simultaneously forward data packets, flooding speed can be accelerated. We investigate the new problems when both a mobile sink and network coding are integrated with a schedule-tree-based flooding scheme. We model the process to deliver multiple encoded packets from the source to a node as a Markov process, and we theoretically investigate how to minimize the total expected flooding delay with a mobile sink by considering all the possible traveling orders of the branches. We propose a heuristic algorithm to obtain an appropriate trajectory by accounting for the effect of traveling delay of the mobile sink. Simulation results show that our proposed flooding scheme with a mobile sink can significantly reduce the flooding delay.

As the foundation of routing, topology control should minimize the interference among nodes, and increase the network capacity. With the development of mobile ad hoc networks (MANETs), there is a growing requirement of quality of service (QoS) in terms of delay. In order to meet the delay requirement, it is important to consider topology control in delay constrained environment, which is contradictory to the objective of minimizing interference. In this paper, we focus on the delay-constrained topology control problem, and take into account delay and interference jointly. We propose a cross-layer distributed algorithm called interference-based topology control algorithm for delay-constrained (ITCD) MANETs with considering both the interference constraint and the delay constraint, which is different from the previous work. The transmission delay, contention delay and the queuing delay are taken into account in the proposed algorithm. Moreover, the impact of node mobility on the interference-based topology control algorithm is investigated and the unstable links are removed from the topology. The simulation results show that ITCD can reduce the delay and improve the performance effectively in delay-constrained mobile ad hoc networks.

In asynchronous wireless sensor networks, finding an optimal candidate set can significantly improve the efficiency of an opportunistic routing protocol by balancing the one-hop cost from a sender to a relay and the remaining cost from the relay to the sink. A larger candidate set indicates smaller one-hop waiting cost to wait for a candidate to wake up, however, a packet is more likely to deviate from the shortest path, which will induce larger remaining cost. Remaining cost of each candidate can be used to construct a coordinate system. Neighbors of a node with high coordinate value is excluded from the candidate set, thus, we can select the next-hop relay by simply taking the first wake-up candidate. In this paper, we propose a distributed algorithm which is similar to the Bellman-Ford algorithm to construct a coordinate system and find an optimal candidate set for each node. Moreover, we propose an algorithm which induces acceptable overhead to maintain the coordinate system and each node's optimal candidate set. We also use simulation results to show that our proposed opportunistic routing protocol is more efficient than conventional strategies in terms of delay, energy efficiency, and miss ratio.

Broadcast is one of the most fundamental services in WSNs. In asynchronous WSNs which every node has different wake/sleep schedule, the broadcast protocol may be achieved by multiple unicasts. Moreover, in the duty-unaware asynchronous networks which the nodes do not know when their neighbors wake up, it is difficult to realize the broadcast protocol. In our paper, we propose a new broadcast protocol to reduce the data redundancy and total transmission energy under the condition of covering the whole networks. We first set up a global optimal transmission cost tree, then based on this tree we set up a local delay coordinate for each node to decrease the total waiting delay. The proposed protocol combines the global optimization and local optimization to balance the transmission cost and waiting cost in broadcasting. The experiment evaluation shows that our proposed protocol can be more efficient in duty-unaware WSNs than the ADB protocol.

This paper discusses the development of novel switching devices using embedded microporcessing systems. It first reviews on basic Ethernet switch operations and the spanning tree protocol (STP). It then gives a brief analysis on Linux STP implementation and bridge configuration tools. The aim of the analysis is to lay a foundation for latter discussion on how novel switching devices can be developed based on the bridging codes. To facilitate the discussion, a new kind of Ethernet switches (with modified STP operations) proposed by the authors of this paper is used as an example. This practical example demonstrates how kernel programming on Linux and some modifications on configuration tools can be made to develop new switching devices. Experiments on the newly developed switches are also reported. The results show that these new switches can provide better security for STP networks. At the end of the paper, discussion on how to port the design to embedded miscroprocessing systems is also given.