Search

Knowledge Graph (KG) has attracted more and more companies' attention for its ability to connect different types of data in meaningful ways and support rich data services. However, due to privacy concerns, different companies cannot share their own KGs with each other. Such data isolation problem limits the performance of KG and prevents its further development. Therefore, how to let multiple parties conduct KG-related tasks collaboratively on the basis of privacy protection becomes an important research question to answer. In this paper, to fill this gap, we summarize the open problems for privacy-preserving KG in the data isolation setting and propose possible solutions for them. Specifically, we summarize the open problems in privacy-preserving KG from four aspects, i.e., merging, query, representation, and completion. We present these problems in detail and propose possible technical solutions for them, along with the datasets, evaluation methods, and future research directions. We also provide three privacy-preserving KG application scenarios. [ABSTRACT FROM AUTHOR]

Graph collaborative filtering (GCF) has achieved exciting recommendation performance with its ability to aggregate high-order graph structure information. Recently, contrastive learning (CL) has been incorporated into GCF to alleviate data sparsity and noise issues. However, most of the existing methods employ random or manual augmentation to produce contrastive views that may destroy the original topology and amplify the noisy effects. We argue that such augmentation is insufficient to produce the optimal contrastive view, leading to suboptimal recommendation results. In this article, we proposed a Learnable Model Augmentation Contrastive Learning (LMACL) framework for recommendation, which effectively combines graph-level and node-level collaborative relations to enhance the expressiveness of collaborative filtering (CF) paradigm. Specifically, we first use the graph convolution network (GCN) as a backbone encoder to incorporate multi-hop neighbors into graph-level original node representations by leveraging the high-order connectivity in user-item interaction graphs. At the same time, we treat the multi-head graph attention network (GAT) as an augmentation view generator to adaptively generate high-quality node-level augmented views. Finally, joint learning endows the end-to-end training fashion. In this case, the mutual supervision and collaborative cooperation of GCN and GAT achieves learnable model augmentation. Extensive experiments on several benchmark datasets demonstrate that LMACL provides a significant improvement over the strongest baseline in terms of Recall and NDCG by 2.5%–3.8% and 1.6%–4.0%, respectively. Our model implementation code is available at https://github.com/LiuHsinx/LMACL. [ABSTRACT FROM AUTHOR]

Sequential recommendation aims to predict the next item of interest to users based on their historical behavior data. Usually, users' global and local preferences jointly affect the final recommendation result in different ways. Most existing works use transformers to globally model sequences, which makes them face the dilemma of quadratic computational complexity when dealing with long sequences. Moreover, the scope setting of the user's local preference is usually static and single, and cannot cover richer multi-level local semantics. To this end, we proposed a parallel architecture for capturing global representation and Multi-granularity Local dependencies with MLP for sequential Recommendation (MLM4Rec). For global representation, we utilize modified MLP-Mixer to capture global information of user sequences due to its simplicity and efficiency. For local representation, we incorporate convolution into MLP and propose a multi-granularity local awareness mechanism for capturing richer local semantic information. Moreover, we introduced a weight pooling method to adaptively fuse local-global representations instead of directly concatenation. Our model has the advantages of low complexity and high efficiency thanks to its simple MLP structure. Experimental results on three public datasets demonstrate the effectiveness of our proposed model. Our code is available here1. [ABSTRACT FROM AUTHOR]

Metal halide perovskites are extensively utilized in photoluminescence and photoelectric detection, owing to their modifiable band gap, elevated quantum yield, high absorption coefficient, superior charge transport properties, versatile chemical synthesis, and cost-effective manufacturing. Among the perovskite variants, bismuth–halide perovskites have emerged as a promising lead–halide alternative, garnering increasing interest for crafting non-toxic perovskites in optoelectronic applications. Nonetheless, the rapid and straightforward production of high-quality single crystals remains a challenge. This paper introduces an enhanced antisolvent crystallization method for fabricating high-quality Cs3Bi2Br9 single crystals measuring 3 × 3 mm. These crystals exhibit a remarkably low half-peak width of the X-ray diffraction swing curve at just 0.048°, maintaining stability without phase decomposition for 180 days at room temperature under air exposure. The photodetector constructed involves an aluminum electrode deposited onto the crystal surface. Utilizing 375 nm ultraviolet light as the excitation source, the device demonstrates a swift electrical response (sensitivity: 8.41 × 102), achieving the rise and decay times of the Cs3Bi2Br9 single crystal detector of 96 and 80 milliseconds, respectively. This research holds significant promise for advancing perovskite crystal preparation and photodetection techniques, particularly in the realm of bismuth-based perovskites. [ABSTRACT FROM AUTHOR]

Cs3Cu2Cl5 is unstable owing to ionic migration and lattice decomposition in the atmosphere. In addition, obtaining large bulk single crystals of Cs3Cu2Cl5 is challenging. Herein, a novel strategy is proposed for synthesizing a Cs3Cu2Cl5 single crystal that exhibits excellent crystallinity and photoluminescence (PL) properties using an antisolvent‐assisted method. Na+ is doped into the Cs3Cu2Cl5 lattice to replenish the lattice defects caused by chlorine vacancies, thus leading to stronger chemical interactions between Cu+ and Cl− ions. Moreover, Na+ doping circumvents ionic migration and lattice decomposition, thereby enhancing the PL intensity and maintaining the long‐term stability of Cs3Cu2Cl5 in the atmosphere. Incorporating 10% Na+ into the Cs3Cu2Cl5 lattice enhances the PL intensity by 18%, and the high‐stability PL can maintain more than 48.5% of the PL intensity after 90 d in an atmospheric environment. In addition, a white light‐emitting device (LED) is fabricated using the 10% Na+‐doped Cs3Cu2Cl5 crystal powder and it exhibits a high color‐rendering index (93.7) and correlated color temperature (7120 K). Additionally, it exhibits superior stability, even at a high temperature of 120 °C. Thus, the excellent high‐temperature stability of Cs3Cu2Cl5 can promote its practical application in LED. [ABSTRACT FROM AUTHOR]

Graph Convolution Network (GCN) has been widely applied in recommender systems for its representation learning capability on user and item embeddings. However, GCN is vulnerable to noisy and incomplete graphs, which are common in real world, due to its recursive message propagation mechanism. In the literature, some work proposes to remove the feature transformation during message propagation, but making it unable to effectively capture the graph structural features. Moreover, they model users and items in the Euclidean space, which has been demonstrated to have high distortion when modeling complex graphs, further degrading the capability to capture the graph structural features and leading to sub-optimal performance. To this end, in this paper, we propose a simple yet effective Quaternion-based Graph Convolution Network (QGCN) recommendation model. In the proposed model, we utilize the hyper-complex Quaternion space to learn user and item representations and feature transformation to improve both performance and robustness. Specifically, we first embed all users and items into the Quaternion space. Then, we introduce the quaternion embedding propagation layers with quaternion feature transformation to perform message propagation. Finally, we combine the embeddings generated at each layer with the mean pooling strategy to obtain the final embeddings for recommendation. Extensive experiments on three public benchmark datasets demonstrate that our proposed QGCN model outperforms baseline methods by a large margin. [ABSTRACT FROM AUTHOR]

Detecting fake reviews can help customers make better purchasing decisions and maintain a positive online business environment. In recent years, pre-trained language models have significantly improved the performance of natural language processing tasks. These models are able to generate different representation vectors for each word in different contexts, thus solving the challenge of multiple meanings of a word, which traditional word vector methods such as Word2Vec cannot solve, and, therefore, better capturing the text's contextual information. In addition, we consider that reviews generally contain rich opinion and sentiment expressions, while most pre-trained language models, including BERT, lack the consideration of sentiment knowledge in the pre-training stage. Based on the above considerations, we propose a new fake review detection model based on a pre-trained language model and convolutional neural network, which is called BSTC. BSTC considers BERT, SKEP, and TextCNN, where SKEP is a pre-trained language model based on sentiment knowledge enhancement. We conducted a series of experiments on three gold-standard datasets, and the findings illustrate that BSTC outperforms state-of-the-art methods in detecting fake reviews. It achieved the highest accuracy on all three gold-standard datasets—Hotel, Restaurant, and Doctor—with 93.44%, 91.25%, and 92.86%, respectively. [ABSTRACT FROM AUTHOR]

Beta zeolite, a crystal material with a three-dimensional twelve-ring cross-channel structure, has many advantages, such as high Brønsted acid concentration, high Si/Al ratio, thermal/hydrothermal stability, and large surface area. Due to these advantages, beta zeolite shows excellent catalytic performance in petroleum refining and petrochemical processes. However, traditionally microporous beta zeolite has strong steric hindrance and diffusion restrictions, which hinder large molecules from passing through its internal channels. In addition, carbon deposition occurs, resulting in catalyst deactivation. The main strategy to solve this problem is to prepare nanosized or hierarchical beta zeolites, which allow for large molecule conversion and shortening diffusion pathways. Therefore, researchers have explored different synthesis strategies to prepare beta zeolite with different particle sizes and porosities to obtain better zeolite catalysts. This paper briefly describes the recent research progress in the preparation of nanosized and hierarchical beta zeolite. Additionally, the mechanisms of various preparation methods, structural characteristics, and applications of the materials are introduced in detail. Furthermore, the main problems existing in its industrial application are describing by comparing the advantages and disadvantages of the different methods to prepare optimally nanosized and hierarchical zeolite to meet the requirements of industrial development. [ABSTRACT FROM AUTHOR]

The goal of aspect-based sentiment analysis (ABSA) is to identify the sentiment polarity of specific aspects in a context. Recently, graph neural networks have employed dependent tree syntactic information to assess the link between aspects and contextual words; nevertheless, most of this research has neglected phrases that are insensitive to syntactic analysis and the effect between various aspects in a sentence. In this paper, we propose a dual-channel edge-featured graph attention networks model (AS-EGAT), which builds an aspect syntactic graph by enhancing the contextual syntactic dependency representation of key aspect words and the mutual affective relationship between various aspects in the context and builds a semantic graph through the self-attention mechanism. We use the edge features as a significant factor to determine the weight coefficient of the attention mechanism to efficiently mine the edge features of the graph attention networks model (GAT). As a result, the model can connect important sentiment features of related aspects when dealing with aspects that lack obvious sentiment expressions, pay close attention to important word aspects when dealing with multiple-word aspects, and extract sentiment features from sentences that are not sensitive to syntactic dependency trees by looking at semantic features. Experimental results show that our proposed AS-EGAT model is superior to the current state-of-the-art baselines. Compared with the baseline models of LAP14, REST15, REST16, MAMS, T-shirt, and Television datasets, the accuracy of our AS-EGAT model increased by 0.76%, 0.29%, 0.05%, 0.15%, 0.22%, and 0.38%, respectively. The macro-f1 score increased by 1.16%, 1.16%, 1.23%, 0.37%, 0.53%, and 1.93% respectively. [ABSTRACT FROM AUTHOR]

Zeolite Y, as a solid acid catalyst with excellent performance, is a landmark in petroleum refining and chemical industry production–especially in catalytic cracking reactions. Improving the SAR of Y zeolite, enriching its pore structure, and modifying it with heteroatoms can realize the multifunctional catalysis of Y zeolite, improve the application value of it, and then meet the demands of petroleum refining. In this review, the synthesis of Y zeolites with high SAR, multistage pores, and heteroatom modification is summarized. [ABSTRACT FROM AUTHOR]

Recently, researches on dynamic decision-making based on streaming data are in full swing. As an indispensable technology for data management and analysis, indexing methods are also evolving. The indexing paradigm named learned index has been proposed to replace the traditional B-tree family in some scenarios. It has been proved that learned indexes can provide higher lookup efficiency and lower storage cost overhead than traditional indexes. Usually, learned indexes assume that the data is static or at least the data distribution is unchanged. However, the streaming scenarios break the strong assumption. This paper presents a learned index scheme for streaming scenarios (LIFOSS for short), where the workloads insert, delete, and query data arbitrarily. Precisely, LIFOSS consists of three parts: a) an adaptive packed-memory array which stores data and handles updates with lower bound of performance guaranteed; b) a middle-layer model group, used to fit the cumulative distribution function of data; c) a feedback mechanism designed to update parameters of the model group above in real-time locally. Extensive experiments on two public datasets show that LIFOSS performs better than the state-of-the-art dynamic learned index method. LIFOSS reduces the lookup latency by at least 6 % , and its dynamic performance is more stable, requiring only a tiny amount of extra space. [ABSTRACT FROM AUTHOR]

With the explosion of GPS-enabled smartphones and social media platforms, geo-social networks are increasing as tools for businesses to promote their products or services. Influence maximization, which aims to maximize the expected spread of influence in the networks, has drawn increasing attention. However, most recent work tries to study influence maximization by only considering geographic distance, while ignoring the influence of users’ spatio-temporal behavior on information propagation or location promotion, which can often lead to poor results. To relieve this problem, we propose a Similarity-aware Influence Maximization (SIM) model to efficiently maximize the influence spread by taking the effect of users’ spatio-temporal behavior into account, which is more reasonable to describe the real information propagation. We first calculate the similarity between users according to their historical check-ins, and then we propose a Propagation to Consumption (PTC) model to capture both online and offline behaviors of users. Finally, we propose two greedy algorithms to efficiently maximize the influence spread. The extensive experiments over real datasets demonstrate the efficiency and effectiveness of the proposed algorithms. [ABSTRACT FROM AUTHOR]

Narrow passages in free space pose great challenges to many sampling-based motion planners. In problems with high-dimensional configuration spaces (C-spaces), narrow passages in free space (C-free) can be hard to find and navigate and, sometimes, even counterintuitive. In this article, we present an algorithm to construct cones of available instantaneous velocities (“free velocity cones” or, briefly, “free cones”) on the boundary of C-free to facilitate finding and navigating narrow passages. This is accomplished by first developing free cones and associated measures of local C-space narrowness for a single rigid link. These results are then extended to kinematic chains (open and closed) of arbitrary degrees of freedom. It turns out that the degeneracy of the free cones dictates the existence of C-space narrow passages, and the locations and orientations of the links in the workspace relate intimately to the corresponding connected component of C-free. This observation leads us to a modified probabilistic roadmap (M-PRM) algorithm and a modified rapidly exploring random vine (M-RRV) algorithm of combining the enumeration of topological components and random sampling of the free cones. Experimental results from applying our algorithms to several challenging examples show that our new algorithms are more efficient than some variants of the PRM algorithms, and several variants of the rapidly exploring random tree (RRT) algorithms, such as RRT-CONNECT and RRV. [ABSTRACT FROM AUTHOR]

The sequential fluorination-alkaline treatment protocol has been applied for the tailoring of siliceous ZSM-5 zeolite. The original spontaneous growth of mesoporosity in alkaline medium is altered due to the antecedent fluorination step. The outcome is demonstrated by the apparent delay in the mesoporosity growth, whose essential duration for the well-defined mesoporosity is therefore extended from 30 min to 60 min. A low fluorination level decelerates the mesoporosity growth, whereas a high fluorination level enables the achievement of the mesoporosity. These impacts are closely linked with the alteration to the states of Al sites as the function of fluorination level. Compared to the states of Al sites in the pristine and steamed zeolites, the electronic and steric consequences on the environment of Al species by fluorination is proposed for the interplay with the alkaline medium for the mesoporosity growth. [ABSTRACT FROM AUTHOR]

With the ubiquity of smart devices, Spatial Crowdsourcing (SC) has emerged as a new transformative platform that engages mobile users to perform spatio-temporal tasks by physically traveling to specified locations. Thus, various SC techniques have been studied for performance optimization, among which one of the major challenges is how to assign workers the tasks that they are really interested in and willing to perform. In this paper, we propose a novel preference-aware spatial task assignment system based on workers’ temporal preferences, which consists of two components: History-based Context-aware Tensor Decomposition (HCTD) for workers’ temporal preferences modeling and preference-aware task assignment. We model workers’ preferences with a three-dimension tensor (worker-task-time). Supplementing the missing entries of the tensor through HCTD with the assistant of historical data and other two context matrices, we recover workers’ preferences for different categories of tasks in different time slots. Several preference-aware individual task assignment algorithms are then devised, aiming to maximize the total number of task assignments at every time instance, in which we give higher priorities to the workers who are more interested in the tasks. In order to make our proposed framework applicable to more scenarios, we further optimize the original framework by proposing strategies to allow each task to be assigned to a group of workers such that the task can be completed by these workers simultaneously, wherein workers’ tolerable waiting time, consensus, and tasks’ rewards are taken into consideration. We conduct extensive experiments using a real dataset, verifying the practicability of our proposed methods. [ABSTRACT FROM AUTHOR]

Online ride-hailing platforms have reduced significantly the amounts of the time that taxis are idle and that passengers spend on waiting. As a key component of these platforms, the fleet management problem can be naturally modeled as a Markov Decision Process, which enables us to use the deep reinforcement learning. However, existing studies are proposed based on simplified problem settings that fail to model the complicated supply-dynamics and restrict the performance in the real traffic environment. In this article, we propose a supply-demand-aware deep reinforcement learning algorithm for taxi dispatching, where we use a deep Q-network with action sampling policy, called AS-DQN, to learn an optimal dispatching policy. Furthermore, we utilize a dueling network architecture, called AS-DDQN, to improve the performance of AS-DQN. Extensive experiments on real-world datasets offer insight into the performance of our model and show that it is capable of outperforming the baseline approaches. [ABSTRACT FROM AUTHOR]

Massive trajectory data stem from the prevalence of equipment-supporting GPS and wireless communication technology. Especially, activity trajectory from Location-based Social Network (LBSN) endows traditional trajectory data with additional user semantic activities, e.g., visiting work/home/entertainment places. Measuring the similarity between activity trajectories is to compare their proximity in multiple dimensions such as time, location, and semantics. In this way, we can mine implicit user preference and apply it to route planning, POI recommendation or any other online tasks. The key challenge of comparing activity trajectories (i.e., computing their similarity) lies in two aspects. One is the uneven sampling rate in both time and space. The other is the discrepancy of individual activities. Previous effort alleviates the issue of uneven sampling rate via trajectory complements, which is limited to spatial-temporal information. In this paper, we propose to learn a representation for one activity trajectory by jointly considering the spatio-temporal characteristics and the activity semantics. The similarity of two trajectories is computed by weighting individual trajectory points and contextual features with multi-level attention mechanisms. In specific, we propose a point-level and feature-level attention mechanism to adaptively select critical elements and contextual factors for learning trajectory representation. Our proposed approach, called At2vec, demonstrates better performance than existing baselines in extensive experimental evaluation on real trajectory databases. [ABSTRACT FROM AUTHOR]

There is relatively little research on deep learning for anomaly detection within the field of deep learning. Existing deep anomaly detection methods focus on the learning of feature reconstruction, but such methods mainly learn new feature representations, and the new features do not fully reflect the original features, leading to inaccurate anomaly scores; in addition, there is an end-to-end deep anomaly detection algorithm, but the method cannot accurately obtain a reference score that matches the data themselves. In addition, in most practical scenarios, the data are unlabeled, and there exist some datasets with labels, but the confidence and accuracy of the labels are very low, resulting in inaccurate results when put into the model, which makes them often designed for unsupervised learning, and thus in such algorithms, the prior knowledge of known anomalous data is often not used to optimize the anomaly scores. To address the two problems raised above, this paper proposes a new anomaly detection model that learns anomaly scores mainly through a variational deviation network (i.e., not by learning the reconstruction error of new features, distance metrics, or random generation, but by learning the normal distribution of normal data). In this model, we force the anomaly scores to deviate significantly from the normal data by a small amount of anomalous data and a reference score generated by variational self-encoding. The experimental results in multiple classes of data show that the new variational deviation network proposed in this paper has higher accuracy among the mainstream anomaly detection algorithms. [ABSTRACT FROM AUTHOR]

Integrating social networks as auxiliary information shows effectiveness in improving the performance for a recommendation task. Typical models usually characterize the user trust relationship as a binary adjacent matrix derived from a social graph, which basically only incorporates neighborhood interactions, and then encodes the trust values of different individuals with the same value. Such methods fail to capture the implicit high-order relations hidden under a graph structure, and thereby ignore the impact of indirect influencers. To address the aforementioned problems, we present an I mplicit T rust R elation-A ware model (ITRA) based on Variational Auto-Encoder (VAE). ITRA adopts an attention module to feed the weighted trust embedding information into an inherited non-linear VAE structure. In this sense, ITRA could provide recommendations by reconstructing a non-binary adjacency social matrix with implicit high-order interactions from both indirect key opinion leaders and explicit connections from neighbors. The extensive experiments conducted on three datasets illustrate that ITRA could achieve a better performance compared to the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Graph Pattern based Node Matching (GPNM) has been proposed to find all the matches of the nodes in a data graph GD based on a given pattern graph GP. GPNM has been increasingly adopted in many applications such as group finding and expert recommendation, in which data graphs are frequently updated over time. Moreover, many typical pattern graphs frequently and repeatedly appear in users’ queries in a short period of time, e.g., social graph searches on Facebook. To deliver a GPNM result in such applications, the existing GPNM methods have to perform an incremental GPNM procedure for each of the updates in the data graph, which is computationally expensive. To address this problem, in this paper, we first analyze the elimination relationships between multiple updates in GD and the hierarchical structure between these elimination relationships. Then, we generate an Elimination Hierarchy Tree (EH-Tree) to index the elimination relationships and propose an EH-Tree based GPNM method, called EH-GPNM, considering the elimination relationships between multiple updates in GD. EH-GPNM first delivers the GPNM result of an initial query, and then delivers the GPNM result of a subsequent query, based on the initial GPNM result and the multiple updates of GD that occur between those two queries. The experimental results on five real-world social graphs demonstrate that our proposed EH-GPNM is much more efficient than the state-of-the-art GPNM methods. [ABSTRACT FROM AUTHOR]

This article aims to extend the theory of Lie subgroups and symmetric subspaces for studying an important class of submanifolds of the special Euclidean group SE(3) whose tangent space at each point on the submanifold relates to that at the identity by an adjoint map. These submanifolds, which we call adjoint-invariant submanifolds in this article, are known in the literature as persistent submanifolds, since they are strictly related to the concept of persistent screw systems. The difference is that in this article, just as Lie subgroups and symmetric subspaces, we put forward adjoint-invariant submanifolds as independent geometric objects from mechanisms and their associated local screw systems. Adjoint invariance relaxes the strict left and right invariance of Lie subgroups and the reflective invariance of symmetric subspaces by allowing generic moving reference frame in the aforementioned adjoint map. It turns out such adjoint invariance can be studied under the framework of distributions on manifolds, which allows us to explore global geometric properties of adjoint-invariant submanifolds. We classify adjoint-invariant submanifolds into reflective-type and product-type submanifolds and derive the conditions for their adjoint invariance. We then propose geometric methods and algorithms for synthesizing the kinematic generators for reflective-type submanifolds, as demonstrated with a number of examples. [ABSTRACT FROM AUTHOR]

The pervasiveness of GPS-enabled devices and wireless communication technologies results in massive trajectory data, incurring expensive cost for storage, transmission, and query processing. To relieve this problem, in this paper we propose a novel framework for compressing trajectory data, REST (Reference-based Spatio-temporal trajectory compression), by which a raw trajectory is represented by concatenation of a series of historical (sub-)trajectories (called reference trajectories) that form the compressed trajectory within a given spatio-temporal deviation threshold. In order to construct a reference trajectory set that can most benefit the subsequent compression, we propose three kinds of techniques to select reference trajectories wisely from a large dataset such that the resulting reference set is more compact yet covering most footprints of trajectories in the area of interest. To address the computational issue caused by the large number of combinations of reference trajectories that may exist for resembling a given trajectory, we propose efficient greedy algorithms that run in the blink of an eye and dynamic programming algorithms that can achieve the optimal compression ratio. Compared to existing work on trajectory compression, our framework has few assumptions about data such as moving within a road network or moving with constant direction and speed, and better compression performance with fairly small spatio-temporal loss. In addition, by indexing the reference trajectories directly with an in-memory R-tree and building connections to the raw trajectories with inverted index, we develop an extremely efficient algorithm that can answer spatio-temporal range queries over trajectories in their compressed form. Extensive experiments on a real taxi trajectory dataset demonstrate the superiority of our framework over existing representative approaches in terms of both compression ratio and efficiency. [ABSTRACT FROM AUTHOR]

Point-of-interest (POI) recommendation has great significance in helping users find favorite places from a large number of candidate venues. One challenging in POI recommendation is to effectively exploit geographical information since users usually care about the physical distance to the recommended POIs. Though spatial relevance has been widely considered in recent recommendation methods, it is modeled only from the POI perspective, failing to capture user personalized preference to spatial distance. Moreover, these methods suffer from a diversity-deficiency problem since they are often based on collaborative filtering which always favors popular POIs. To overcome these problems, we propose in this article a personalized geographical influence modeling method called PGIM, which jointly learns users' geographical preference and diversity preference for POI recommendation. Specifically, we model geographical preference from three aspects: user global tolerance, user local tolerance, and spatial distance. We also extract user diversity preference from interactions among users for diversity-promoting recommendation. Experimental results on three real-world datasets demonstrate the superiority of PGIM. [ABSTRACT FROM AUTHOR]

Purpose: Unlike the normal organ segmentation task, automatic tumor segmentation is a more challenging task because of the existence of similar visual characteristics between tumors and their surroundings, especially on computed tomography (CT) images with severe low contrast resolution, as well as the diversity and individual characteristics of data acquisition procedures and devices. Consequently, most of the recently proposed methods have become increasingly difficult to be applied on a different tumor dataset with good results, and moreover, some tumor segmentors usually fail to generalize beyond those datasets and modalities used in their original evaluation experiments. Methods: In order to alleviate some of the problems with the recently proposed methods, we propose a novel unified and end-to-end adversarial learning framework for automatic segmentation of any kinds of tumors from CT scans, called CTumorGAN, consisting of a Generator network and a Discriminator network. Specifically, the Generator attempts to generate segmentation results that are close to their corresponding golden standards, while the Discriminator aims to distinguish between generated samples and real tumor ground truths. More importantly, we deliberately design different modules to take into account the well-known obstacles, e.g., severe class imbalance, small tumor localization, and the label noise problem with poor expert annotation quality, and then use these modules to guide the CTumorGAN training process by utilizing multi-level supervision more effectively. Results: We conduct a comprehensive evaluation on diverse loss functions for tumor segmentation and find that mean square error is more suitable for the CT tumor segmentation task. Furthermore, extensive experiments with multiple evaluation criteria on three well-established datasets, including lung tumor, kidney tumor, and liver tumor databases, also demonstrate that our CTumorGAN achieves stable and competitive performance compared with the state-of-the-art approaches for CT tumor segmentation. Conclusion: In order to overcome those key challenges arising from CT datasets and solve some of the main problems existing in the current deep learning-based methods, we propose a novel unified CTumorGAN framework, which can be effectively generalized to address any kinds of tumor datasets with superior performance. [ABSTRACT FROM AUTHOR]

Address matching is a crucial task in various location-based businesses like take-out services and express delivery, which aims at identifying addresses referring to the same location in address databases. It is a challenging one due to various possible ways to express the address of a location, especially in Chinese. Traditional address matching approaches relying on string similarities and learning matching rules to identify addresses referring to the same location, could hardly solve the cases with redundant, incomplete or unusual expression of addresses. In this paper, to learn the geographical semantic representations for address strings, we novelly propose to get rich contexts for addresses from the Web through Web search engines, which could strongly enrich the semantic meaning of addresses that could be learned. Apart from that, we propose a two-stage geographical address representation learning model for address matching. In the first stage, we propose to use an encode-decoder architecture to learn the semantic vector representation for each address string where an up-sampling and sub-sampling strategy is applied to solve the problem of address redundancy and incompleteness. The attention mechanism is also applied to the model to highlight important features of addresses in their semantic representations. And in the second stage, we construct a single large graph from the corpus, which contains address elements and addresses as nodes, and the edges between nodes are built by word co-occurrence information to learn embedding representations for all the nodes on the graph. Our empirical study conducted on two real-world address datasets demonstrates that our approach greatly improves both precision (up to 8%) and recall (up to 12%) of the state-of-the-art existing methods. [ABSTRACT FROM AUTHOR]

With the ubiquity of mobile devices and wireless networks, Spatial Crowdsourcing (SC) has earned considerable importance and attention as a new strategy of problem-solving. Tasks in SC have location constraints and workers need to move to certain locations to perform them. Current studies mainly focus on maximizing the benefits of the SC platform. However, user average waiting time, which is an important indicator of user experience, has been overlooked. To enhance user experience, the SC platform needs to collect lots of data from both workers and users. During this process, the private information may be compromised if the platform is not trustworthy. In this paper, we first define user experience-driven secure task assignment problem and propose two privacy-preserving online task assignment strategies to minimize the average waiting time. We securely construct an encrypted bipartite graph to protect private data. Based on this encrypted graph, we propose a secure Kuhn-Munkres algorithm to realize task assignment without privacy disclosure. Theoretical analysis shows the security of our approach and experimental results demonstrates its efficiency and effectiveness. [ABSTRACT FROM AUTHOR]

Graph pattern matching is to find the subgraphs matching the given pattern graphs. In complex contextual social networks, considering the constraints of social contexts like the social relationships, the social trust, and the social positions, users are interested in the top-K matches of a specific node (denoted as the designated node) based on a pattern graph, rather than the entire set of graph matching. This inspires the conText-Aware Graph pattern-based top-K designated node matching (TAG-K) problem, which is NP-complete. Targeting this challenging problem, we propose a recurrent neural network- (RNN-) based Monte Carlo Tree Search algorithm (RN-MCTS), which automatically balances exploring new possible matches and extending existing matches. The RNN encodes the subgraph and maps it to a policy which is used to guide the MCTS. The experimental results demonstrate that our proposed algorithm outperforms the state-of-the-art methods in terms of both efficiency and effectiveness. [ABSTRACT FROM AUTHOR]

Collaborative Web services QoS prediction (CQoSP) has been proved to be an effective tool to predict unknown QoS values of services. Recently a number of efforts have been made in this area, focusing on improving the accuracy of prediction. In this paper, we consider a novel kind of CQoSP, shared CQoSP, where multiple parties share their data with each other in order to provide more accurate prediction than a single party could do. To encourage data sharing, we propose a privacy-preserving framework which enables shared collaborative QoS prediction without leaking the private information of the involved party. Our framework is based on differential privacy, a rigorous and provable privacy model. We conduct extensive experiments on a real Web services QoS dataset. Experimental results show the proposed framework increases prediction accuracy while ensuring the privacy of data owners. [ABSTRACT FROM AUTHOR]

Temporal path discovery in dynamic graphs is significant in many applications, such as the trip planning in transportation networks, and the disease progression tracking in gene networks. Attributed Dynamic Graph (ADG) contains multiple value-changed attributes on edges, such as the speed of a bus between two stations, and the price of the bus ticket in different time periods. The traditional methods for temporal path discovery in ADGs assume users only consider a single constraint on the attributes in their models, such as finding the fast route to reach the destination under the constraint of a given budget, which makes the path discovery problem simple though it still suffers expensive time cost. However, such an assumption is too strict in real applications, where users can specify multiple constraints, such as finding the fast route under the constraints of a given budget and the total number of stopovers. In such a situation, the temporal path discovery becomes more complicated as it subsumes the classical NP-Complete Multi-Constraint Path (MCP) problem, and thus the traditional methods cannot work for finding a new type of Temporal Path with Multiple Constraints (TPMC). In this paper, we propose a set of new Two-Pass approximation algorithms to bi-directionally search ADGs to find TPMC results. To the best of our knowledge, our Two-Pass algorithms are the first algorithms to support the discovery of the temporal paths with multiple constraints in ADGs. The experimental results on 12 real-world dynamic graphs demonstrate that our algorithms outperform the state-of-the-art methods in both efficiency and effectiveness. [ABSTRACT FROM AUTHOR]

Publication recommendation is an interesting but challenging research problem. Most existing studies only use partial information of papers' contents, reference network or co-author relationship, which leads to an unsatisfied recommendation result. In this study, we propose a novel hybrid publication recommendation approach using compound information which retrieves top-K most relevant papers from a publication depository for a set of user input keywords. Our advantages comparing to the existing methods include: (1) Reaching a better recommendation results by taking the advantages of both content-based recommendation and citation-based recommendation and exploring much richer information of papers in one method; (2) Effectively solving the cold-start problem for new published papers by considering the vitality of papers and the impact factor of venues into the citation network; (3) Saving a large overhead in calculating the content-based similarity between papers and user input keywords by doing paper clustering based on the citation network. Extensive experiments on DBLP and Microsoft Academic datasets demonstrate that PubTeller improves the state-of-the-art methods with 4% in Precision and 4.5% in Recall. [ABSTRACT FROM AUTHOR]

Collaborative Web services QoS prediction has proved to be an important tool to estimate accurately personalized QoS experienced by individual users, which is beneficial for a variety of operations in the service ecosystem, such as service selection, composition and recommendation. While a number of achievements have been attained on the study of improving the accuracy of collaborative QoS prediction, little work has been done for protecting user privacy in this process. In this paper, we propose a privacy-preserving collaborative QoS prediction framework which can protect the private data of users while retaining the ability of generating accurate QoS prediction. We introduce differential privacy, a rigorous and provable privacy model, into the process of collaborative QoS prediction. We first present DPS, a method that disguises a user's observed QoS values by applying differential privacy to the user's QoS data directly. We show how to integrate DPS with two representative collaborative QoS prediction approaches. To improve the utility of the disguised QoS data, we present DPA, another QoS disguising method which first aggregates a user's QoS data before adding noise to achieve differential privacy. We evaluate the proposed methods by conducting extensive experiments on a real world Web services QoS dataset. Experimental results show our approach is feasible in practice. [ABSTRACT FROM AUTHOR]

Summary: Estimating the robot state within a known map is an essential problem for mobile robot; it is also referred to "localization". Even LiDAR-based localization is practical in many applications, it is difficult to achieve global localization with LiDAR only for its low-dimension feedback, especially in environments with repetitive geometric features. A sensor-fusion-based localization system is introduced in this paper, which has the capability of addressing the global localization problem. Both LiDAR and vision sensors are integrated, making use of the rich information introduced by vision sensor and the robustness from LiDAR. A hybrid grid-map is built for global localization, and a visual global descriptor is applied to speed up the localization convergence, combined with a pose refining pipeline for improving the localization accuracy. Also, a trigger mechanism is introduced to solve kidnapped problem and verify the relocalization result. The experiments under different conditions are designed to evaluate the performance of the proposed approach, as well as a comparison with the existing localization systems. According to the experimental results, our system is able to solve the global localization problem, and the sensor-fusion mechanism in our system has an improved performance. [ABSTRACT FROM AUTHOR]

Online Social Network (OSN) has been used to enhance service provision and service selection, where trust is one of the most important factors for the decision making of service consumers. Thus, a significant and challenging problem is how to effectively and efficiently find those social trust paths that can yield trustworthy trust evaluation results based on the requirements of a service consumer particularly in contextual OSNs which contains social contexts, like social relationships and social trust between participants, and social positions of participants. In this paper, we propose a new concept called Strong Social Graph (SSG), consisting of participants with strong social connections. We also propose an approach to identify SSGs, and propose a novel index method and a graph compression method for SSG. Then based on the compressed SSG and indices, we propose a new efficient and effective approximation algorithm, called SSG-MCBA by adopting the Monte Carlo method and our optimization search strategies. The experiments conducted onto two real social network datasets illustrate that SSG-MCBA greatly outperforms the state-of-the-art method in both efficiency and effectiveness. [ABSTRACT FROM AUTHOR]

The ubiquity of mobile device and wireless networks flourishes the market of spatial crowdsourcing, in which location constrained tasks are sent to workers and expected to be performed in some designated locations. To obtain a global optimal task assignment scheme, the platform usually needs to collect location information of all workers. During this process, there is a significant security concern, that is, the platform may not be trustworthy, so it brings about a threat to workers location privacy. In this paper, to tackle the privacy-preserving task assignment problem, we propose a privacy-preserving reverse auction based assignment model which consists of two key parts. In the first part, we generalize private location to travel cost and protect it by an anonymity based data aggregation protocol. In the second part, we propose a reverse auction task assignment algorithm, which is a truthful incentive mechanism, to encourage workers to offer authentic data. We theoretically show that the proposed model is secure against semi-honest adversaries. Experimental results show that our model is efficient and can scale to real SC applications. [ABSTRACT FROM AUTHOR]

Graph Pattern Matching (GPM) is to find those subgraphs that match a given pattern graph. In many applications, users are interested in the top-k nodes that matches the label of a specific node, (named as the designated node vd) included in a given pattern graph, rather than the entire set of matching. This is called Graph Pattern based Node Matching (GPNM) problem. However, the existing GPM methods for matching the designated node vd in social graphs do not consider the social contexts like the social relationships, the social trust and the social positions which commonly exist in real applications, like the experts recommendation in social graphs, leading to deliver low quality designated nodes. In this paper, we first propose the conText-Aware Graph pattern based Top-K designed nodes finding problem (TAG-K), which involves the NP-Complete Multiple Constrained GPM problem, and thus it is NP-Complete. To address the efficiency and effectiveness issues of TAG-K in large-scale social graphs, we propose two indices, MA-Tree and SSC-Index, which can help efficiently find the Top-K matching. Furthermore, we propose a probabilistic algorithm based on the Monte Carlo Method, called MC-TAG-K. Based on the experimental results on five real social graphs, we have demonstrated that MC-TAG-K outperforms the existing methods in both efficiency and effectiveness. [ABSTRACT FROM AUTHOR]

Web service composition is widely used to extend the function of web services. Different users have different requirements of QoSs (Quality of Services) making them face many problems. The requirement of a special QoS may be a hard requirement or a soft requirement. The hard requirement refers to the QoS which must be satisfied to the user, and the soft one means that the requirement is flexible. This paper tries to solve the service composition problem when there are two kinds of requirements of QoSs. To satisfy various kinds of requirement of the QoS, we propose a composition method based on our proposed framework. We give an analysis from composition models of services and from related QoE (Quality of Experience) of web services. Then, we rank the service candidates and the service requests together. Based on the ranking, a heuristics is proposed for service selection and composition-GLLB (global largest number of service requests first, local best fit service candidate first), which uses "lost value" in the scheduling to denote the QoE. Comparisons are used to evaluate our method. Comparisons show that GLLB reduces the value of NUR (Number of Unfinished service Requests), FV (Failure Value), and AFV (Average Failure Value). [ABSTRACT FROM AUTHOR]

Graph Pattern Matching (GPM) plays a significant role in many real applications, where many applications often need to find Top-K matches of a specific node, (named as the designated node vd) based on a pattern graph, rather than the entire set of matching. However, the existing GPM methods for matching the designated node vd in social graphs do not consider the social contexts like the social relationships, the social trust and the social positions which commonly exist in real applications, like the experts recommendation in social graphs, leading to deliver low quality designated nodes. In this paper, we first propose the conText-Aware Graph pattern based Top-K designed nodes finding problem (TAG-K), which involves the NP-Complete Multiple Constrained GPM problem, and thus it is NP-Complete. To address the efficiency and effectiveness issues of TAG-K in large-scale social graphs, we propose two indices, MA-Tree and SSC-Index, which can help efficiently find the Top-K matching. Furthermore, we propose an approximation algorithm, A-TAG-K. Using real social network datasets, we experimentally verify that A-TAG-K outperforms the existing methods in both efficiency and effectiveness for solving the TAG-K problem. [ABSTRACT FROM AUTHOR]

The goal of this special issue is to solicit high quality, original research contributions on all aspects of trust, privacy and security on crowdsourcing computing, thereby capturing the state of the art and stimulating further developments in the related areas. Paper "A Low Cost and Un-Cancelled Laplace Noise Based Differential Privacy Algorithm for Spatial Decompositions" by Xiaocui Li, Yangtao Wang, Jingkuan Song, Yu Liu, Xinyu Zhang, Ke Zhou, and Chunhua Li studies the security problem of applying differential privacy into spatial docomposition. Paper "Distributed Time-respecting Flow Graph Pattern Matching on Temporal Graphs" by Tianming Zhang, Yunjun Gao, Linshan Qiu, Lu Chen, Qingyuan Linghu, and Shiling Pu explores the problem of distributed graph pattern matching in temporal graphs. [Extracted from the article]

Graph Pattern Matching (GPM) has been used in lots of areas, like biology, medical science, and physics. With the advent of Online Social Networks (OSNs), recently, GPM has been playing a significant role in social network analysis, which has been widely used in, for example, finding experts, social community mining, and social position detection. Given a query which contains a pattern graph $G_Q$ and a data graph $G_D$ , a GPM algorithm finds those subgraphs, $G_M$, that match $G_Q$ in $G_D$ . However, the existing GPM methods do not consider the multiple end-to-end constraints of the social contexts, like social relationships, social trust, and social positions on edges in $G_Q$ , which are commonly found in various applications, such as crowdsourcing travel, social network based e-commerce, and study group selection, etc. In this paper, we first conceptually extend Bounded Simulation to Multi-Constrained Simulation (MCS), and propose a novel NP-Complete Multi-Constrained Graph Pattern Matching (MC-GPM) problem. Then, to address the efficiency issue in large-scale MC-GPM, we propose a new concept called Strong Social Component (SSC), consisting of participants with strong social connections. We also propose an approach to identifying SSCs, and propose a novel index method and a graph compression method for SSC. Moreover, we devise a multithreading heuristic algorithm, called M-HAMC, to bidirectionally search the MC-GPM results in parallel without decompressing graphs. An extensive empirical study over five real-world large-scale social graphs has demonstrated the effectiveness and efficiency of our approach. [ABSTRACT FROM PUBLISHER]

In recent years, social networking sites have been used as a means for a rich variety of activities, such as movie recommendations and product recommendations. In order to evaluate the trust between a truster (i.e., the source) and a trustee (i.e., the target) who have no direct interaction in Online Social Networks (OSNs), the trust network between them that contains important intermediate participants, the trust relations between the participants, and the social context, has an important influence on trust evaluation. Thus, to deliver a reasonable trust evaluation result, before performing any trust evaluation (i.e., trust transitivity), the contextual trust network from a given source to a given target needs to be first extracted from the social network, where constraints on social context should also be considered to guarantee the quality of the extracted networks. However, this problem has been proved to be NP-Complete. Towards solving this challenging problem, we first present a contextual trust-oriented social network structure which takes social contextual impact factors, including trust, social intimacy degree, community impact factor, preference similarity and residential location distance into account. These factors have significant influences on both social interactions between participants and trust evaluation. Then, we present a new concept QoTN (Quality of Trust Network) and propose a social context-aware trust network extraction model. Finally, we propose a Heuristic Social Context-Aware trust Network extraction algorithm (H-SCAN-K) by extending the K-Best-First Search (KBFS) method with several proposed optimization strategies. The experiments conducted on two real datasets illustrate that our proposed model and algorithm outperform the existing methods in both algorithm efficiency and the quality of the extracted trust networks. [ABSTRACT FROM AUTHOR]

Crowdsourcing applications like Amazon Mechanical Turk (AMT) make it possible to address many difficult tasks (e.g., image tagging and sentiment analysis) on the internet and make full use of the wisdom of crowd, where worker quality is one of the most crucial issues for the task owners. Thus, a challenging problem is how to effectively and efficiently select the high quality workers, so that the tasks online can be accomplished successfully under a certain budget. The existing methods on the crowd worker selection problem mainly based on the quality measurement of the crowd workers, those who have to register on the crowdsourcing platforms. With the connect of the OSNs and the crowdsourcing applications, the social contexts like social relationships and social trust between participants and social positions of participants can assist requestors to select one or a group of trustworthy crowdsourcing workers. In this paper, we first present a contextual social network structure and a concept of Strong Social Component (SSC), which emblems a group of workers who have high social contexts values. Then, we propose a novel index for SSC, and a new efficient and effective algorithm C-AWSA to find trustworthy workers, who can complete the tasks with high quality. The results of our experiments conducted on four real OSN datasets illustrate that the superiority of our method in trustworthy worker selection. [ABSTRACT FROM AUTHOR]

A location-aware publish/subscribe (pub/sub) system is gaining more and more interest in both industry and academia with the rapid progress of mobile Internet and the rising popularity of smart-phones. Nowadays, with the booming of E-commerce, OTO (online-to-offline) services are gaining more and more popularity, which results in millions of products with different structured descriptions and locations. To meet this requirement, a pub/sub system should handle subscriptions with location-aware boolean expressions to present users' interests. In this paper, we propose an efficient location-aware pub/sub index for boolean expressions, called RP-trees. RP-trees integrates an R-tree index and a boolean expression index together, can efficiently and simultaneously prune boolean expressions and spatial dimensions. RP-trees is also extensible to support complex environment such as prefix-matching and subscriptions in format of CNF and DNF. Our experimental results show that RP-trees achieves good performance on a synthetic dataset and two real-world datasets (58 city and ebay). [ABSTRACT FROM AUTHOR]

The proliferation of GPS-enabled smart mobile devices enables us to collect a large-scale trajectories of moving objects with GPS tags. While the raw trajectories that only consists of positional information have been studied extensively, many recent works have been focusing on enriching the raw trajectories with semantic knowledge. The resulting data, called activity trajectories, embed the information about behaviors of the moving objects and support a variety of applications for better quality of services. In this paper, we propose a Top-k Spatial Keyword (T kSK) query for activity trajectories, with the objective to find a set of trajectories that are not only close geographically but also meet the requirements of the query semantically. Such kind of query can deliver more informative results than existing spatial keyword queries for static objects, since activity trajectories are able to reflect the popularity of user activities and reveal preferable combinations of facilities. However, it is a challenging task to answer this query efficiently due to the inherent difficulties in indexing trajectories as well as the new complexity introduced by the textual dimension. In this work, we provide a comprehensive solution, including the novel similarity function, hybrid indexing structure, efficient search algorithm and further optimizations. Extensive empirical studies on real trajectory set have demonstrated the scalability of our proposed solution. [ABSTRACT FROM AUTHOR]