Search

International audience; In this paper, we address the problem of learning from im-balanced multi-class datasets in a supervised setting when multiple descriptions of the data-also called views-are available. Each view incorporates various information on the examples, and in particular, depending on the task at hand, each view might be better at recognizing only a subset of the classes. Establishing a sort-of cooperation between the views is needed for all the classes to be equally recognized-a crucial problem particularly for imbalanced datasets. The novelty of our work consists in capitalizing on the complementariness of the views so that each class can be processed by the most appropriate view(s); thus improving the per-class performances of the final classifier. The main contribution of this paper are two ensemble learning methods based on recent theoretical works on the use of the confusion matrix's norm as an error measure, while empirical results show the benefits of the proposed approaches.

In cases of power outages the communication of organizations responsible for recovery work (emergency services, public administration, energy network operators) to the public poses several challenges, primarily the breakdowns of many communication infrastructures and therefore the limitations of the use of classical communication media. This paper surveys technical concepts to support crisis communication during blackouts. Therefore it first investigates the perception and information demands of citizens and communication infrastructures in different scenarios. Furthermore, it analyzes communication infrastructures and their availability in different scenarios. Finally it proposes ‘BlaCom', an ICT-based concept for blackout communication, which addresses the time span between the occurrence of the energy blackout and the possible overload of the mobile phone network. It combines general information with location-specific and setting-specific information, was implemented as a prototype smartphone application and evaluated with 12 potential end users.

PurposeThe best algorithm that was implemented on this Brazilian dataset was artificial immune system (AIS) algorithm. But the time and cost of this algorithm are high. Using asexual reproduction optimization (ARO) algorithm, the authors achieved better results in less time. So the authors achieved less cost in a shorter time. Their framework addressed the problems such as high costs and training time in credit card fraud detection. This simple and effective approach has achieved better results than the best techniques implemented on our dataset so far. The purpose of this paper is to detect credit card fraud using ARO.Design/methodology/approachIn this paper, the authors used ARO algorithm to classify the bank transactions into fraud and legitimate. ARO is taken from asexual reproduction. Asexual reproduction refers to a kind of production in which one parent produces offspring identical to herself. In ARO algorithm, an individual is shown by a vector of variables. Each variable is considered as a chromosome. A binary string represents a chromosome consisted of genes. It is supposed that every generated answer exists in the environment, and because of limited resources, only the best solution can remain alive. The algorithm starts with a random individual in the answer scope. This parent reproduces the offspring named bud. Either the parent or the offspring can survive. In this competition, the one which outperforms in fitness function remains alive. If the offspring has suitable performance, it will be the next parent, and the current parent becomes obsolete. Otherwise, the offspring perishes, and the present parent survives. The algorithm recurs until the stop condition occurs.FindingsResults showed that ARO had increased the AUC (i.e. area under a receiver operating characteristic (ROC) curve), sensitivity, precision, specificity and accuracy by 13%, 25%, 56%, 3% and 3%, in comparison with AIS, respectively. The authors achieved a high precision value indicating that if ARO detects a record as a fraud, with a high probability, it is a fraud one. Supporting a real-time fraud detection system is another vital issue. ARO outperforms AIS not only in the mentioned criteria, but also decreases the training time by 75% in comparison with the AIS, which is a significant figure.Originality/valueIn this paper, the authors implemented the ARO in credit card fraud detection. The authors compared the results with those of the AIS, which was one of the best methods ever implemented on the benchmark dataset. The chief focus of the fraud detection studies is finding the algorithms that can detect legal transactions from the fraudulent ones with high detection accuracy in the shortest time and at a low cost. That ARO meets all these demands.

In the paper, we studied specific features of adaptive testing implementation by means of different methods for the purpose of increasing an objective character of results of such tests and decreasing labor-intensiveness of their creation. We chosen an optimal structure for a hybrid neural network.

The assurance of high quality requirements and system specifications takes a huge effort and requires semantically integrated requirements. Especially when using different representations of requirements, the semantic integration is quite difficult. Today’s requirements management tools integrate requirements only syntactically and, thereby, provide only limited features to support the requirements analyst during quality assurance. The following paper describes a concept for the semantic integration of requirements documented in different representations into one common model. This model allows analysing and measuring the quality of the integrated requirements by applying algorithms. The mentioned integration concept is applied to a use case driven requirements elicitation and analysis process using different representations like several UML diagrams or even template based textual requirements. The foundation to measure the quality of the integrated requirements is explained.

In this paper, a cold standby repairable system consisting of two dissimilar components and one repairman is examined. It was assumed that both Component 1 and Component 2, were not as good as new after repair, and Component 1 had repair priority. Under these assumptions, using the geometric process repair model, a discrete replacement policy N under which the system is replaced when the number of failures of Component 1 reaches N is considered. The problem was to determine an optimal policy N∗ so that the long-run average loss per unit time L( N) of the system was minimized. The expression L( N) was explicitly derived, and the corresponding optimal policy N∗ can be determined by minimizing L( N) with respect to N. A numerical example and the associated MATLAB computer program are given to illustrate the model—s applicability. Finally, conclusions are drawn for this study.

In this paper, we propose an algorithm to achieve both connectivity and coverage for multiple target points in grid sensor networks. First we deal with the local targets block, then we use the least sensors to make all the sensors connected. This algorithm has the O(n) running time complexity and its result is proved closed to the optimal result.

Based on the rotating high-frequency voltage-signal injection technique this paper introduces and compares two different approaches to extract the rotor position information for sensorless low speed control of direct-drive permanent magnet synchronous generator-based wind power system. One of the approaches is to extract the rotor position information from the carrier current of permanent magnet synchronous generator (PMSG) by the traditional bandpass filter. The other one is to use an extended Kalman filter algorithm. The effectiveness of the proposed sensorless low speed control is shown by numerical simulations using MATLAB/Simulink environment. The simulation results show that both control schemes provide an accurate estimate of the position of the PMSG-wind turbine and shaft speed. But extended Kalman filter algorithm improves the efficiency of the high-frequency voltage-signal injection method by eliminate the phase shift introduce by the traditional bandpass filter.

Differentiated service (DiffServ) in combination with multiprotocol label switching (MPLS) is a promising technology in converting the best-effort Internet into a QoS-capable network. This paper describes the design and implementation of key DiffServ components, including classifier, meter/shaper, queue manager, and scheduler, in an MPLS edge router architecture on a network-processor platform. We describe how the DiffServ functionalities were realized and also analyze the performance of the system under different traffic patterns. Various factors that cause performance degradation in the architecture are observed and analyzed. These include receiving processing architecture, transmitting processing architecture, micro-engine processing power, memory access latency, and complexity of each DiffServ component. Among them, we found that traffic classification demands more processing resource and hence is a major factor in limiting the system throughput

Introduction The Research Branch of the Canada Department of Agriculture maintains a Research Station at Lethbridge, Alberta, for the purpose of undertaking research on a broad spectrum of agricultural problems peculiar to the Great Plains area of Canada. Late in 1959 arrangements were made with the Suffield Experimental Station, of the Defence Research Board of Canada, for three scientists to attend a programming course at Suffield, located 130 miles from Lethbridge. Within six months a number of programs were prepared and put into use for routine production runs, and training of additional staff at Lethbridge was commenced. Unfortunately no professional statisticians or mathematicians are employed on the research staff at Lethbridge. The Branch, however, maintains a Statistical Research Service in Ottawa from which assistance may be obtained in handling statistical and mathematical problems. The computer at Suffield is a Stantec Zebra with papertape input and paper-tape and teleprinter output. All programming done by Lethbridge personnel has been in an interpretive system known as Simple Code, which has been described by Ord-Smith (1960). This system, in effect, transforms the Zebra into a computer operating with decimal, floating-point arithmetic and, in the version used at Suffield, with number and instruction stores each of 1108 words. In addition to the floating-point number store there are eight special registers, which allow for indexing operations and a limited amount of fixed-point integer arithmetic.

Industrial development does not take place in all regions at the same speed. This paper seeks explanations for regional differences in the development of manufacturing industry in Central Java. Special attention is paid to the impact of infrastructure (roads, telecommunications, electricity) on industrial development. The spatial distribution of manufacturing industry is analysed by means of both secondary data at the kabupaten level and primary data on 274 firms in various parts of Central Java. In addition to demand side factors, infrastructure does indeed play an important role, but local government bureaucratic procedures for obtaining land and permits are also important.

In this paper, we propose a series of fuzzy temporal protoforms in the framework of the automatic generation of quantitative and qualitative natural language descriptions of processes. The model includes temporal and causal information from processes and attributes, quantifies attributes in time during the process life-span and recalls causal relations and temporal distances between events, among other features. Through integrating process mining techniques and fuzzy sets within the usual Data-to-Text architecture, our framework is able to extract relevant quantitative temporal as well as structural information from a process and describe it in natural language involving uncertain terms. A real use-case in the cardiology domain is presented, showing the potential of our model for providing natural language explanations addressed to domain experts., Comment: Accepted and published in FUZZ-IEEE 2021: https://doi.org/10.1109/FUZZ45933.2021.9494444 Cite as: Y. Fontenla-Seco, A. Bugarin and M. Lama, "Fuzzy Temporal Protoforms for the Quantitative Description of Processes in Natural Language," 2021 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), Luxembourg, Luxembourg, 2021, pp. 1-6, doi: 10.1109/FUZZ45933.2021.9494444

This paper explores how a deeply embedded norm can be used to exert power and control in the governance of distributed infrastructures, such as the Internet. Through the lens of norm conflict, I analyse a case of resistance against the introduction of norms in the transnational governance of Internet routing. In a mixed-methods case study, I examine through an experiment how a community of network operators resists the introduction of data protection and human rights norms in the Internet routing infrastructure. To provide a possible explanation as to how a prevalent and deeply embedded norm enables the resistance to the introduction of new norms, I develop the notion of ‘infrastructural norms’. This concept could help explain why Internet infrastructure governance thus far has largely defied national and international democratic norms. The understanding of how norms are used in infrastructure governance contributes to the study of power in large technical systems.

Cloud Computing is an emerging technology and is being used by more and more IT companies due to its cost saving benefits and ease of use for users. But, it needs to be environment friendly also. Therefore, Green Cloud Computing is the requirement of the today's world. This paper reviews the efforts made by various researchers to make Cloud Computing more energy efficient, to reduce the carbon footprint rate by various approaches and also discusses the concept of virtualization and various approaches which use virtual machines scheduling and migration to show how these can help to make the system more energy efficient. The summary of the main features of the proposed work of different authors that we have reviewed is also presented in it.

This paper discusses the use of EN ISO and EN ISO/IEC standards within the monitoring and reporting regulation and verification regulation. These regulations determine, as their titles might suggest, how (operators of) installations and aircraft operators must determine their emissions, how a verifier must verify emissions, and how verifiers must be accredited. These regulations make extensive use of reference to EN ISO and EN ISO/IEC standards, notably EN ISO/IEC 17011:2004, EN ISO 14065:2013, EN ISO/IEC 17025 and EN ISO 9001. The question arises if these references are in conformity with Article 297 TFEU. A comparative analysis is made with a recent discussion in Dutch jurisprudence and literature about the reference to private standards in Dutch national legislation.

Technological advances, such as IoT devices, cyberphysical systems, smart mobile devices, cloud systems, data analytics, social networks and increased communication capabilities, are making possible to capture, quickly process and analyze huge amounts of data from which to extract information critical for many tasks, such as healthcare and cyber security. Such an intensive use of data raises however major privacy concerns, especially in the mobile application ecosystem. In this paper, we first show different examples of privacy breaches that can arise in such ecosystem. Those examples show that achieving privacy is challenging and there is no unique technique that one can use; rather one must combine different techniques depending also on the intended use of data. We then outline critical challenges towards achieving comprehensive privacy solutions.

Recent statistics in suicide prevention show that people are increasingly posting their last words online and with the unprecedented availability of textual data from social media platforms researchers have the opportunity to analyse such data. Furthermore, psychological studies have shown that our state of mind can manifest itself in the linguistic features we use to communicate. In this paper, we investigate whether it is possible to automatically identify suicide notes from other types of social media blogs in two document-level classification tasks. The first task aims to identify suicide notes from depressed and blog posts in a balanced dataset, whilst the second experiment looks at how well suicide notes can be classified when there is a vast amount of neutral text data, which makes the task more applicable to real-world scenarios. Furthermore we perform a linguistic analysis using LIWC (Linguistic Inquiry and Word Count). We present a learning model for modelling long sequences in two experiment series. We achieve an f1-score of 88.26% over the baselines of 0.60 in experiment 1 and 96.1% over the baseline in experiment 2. Finally, we show through visualisations which features the learning model identifies, these include emotions such as love and personal pronouns.

This paper aims to solve two important issues that frequently occur in existing automatic personality analysis systems: 1. Attempting to use very short video segments or even single frames to infer personality traits; 2. Lack of methods to encode person-specific facial dynamics for personality recognition. Hence, we proposes a novel Rank Loss which utilizes the natural temporal evolution of facial actions, rather than personality labels, for self-supervised learning of facial dynamics. Our approach first trains a generic U-net model that can infer general facial dynamics learned from unlabelled face videos. Then, the generic model is frozen, and a set of intermediate filters are incorporated into this architecture. The self-supervised learning is then resumed with only person-specific videos. This way, the learned filters' weights are person-specific, making them a valuable source for modeling person-specific facial dynamics. We then concatenate the weights of the learned filters as a person-specific representation, which can be directly used to predict the personality traits without needing other parts of the network. We evaluate the proposed approach on both self-reported personality and apparent personality datasets. Besides achieving promising results in personality trait estimation from videos, we show that fusion of tasks reaches highest accuracy, and that multi-scale dynamics are more informative than single-scale dynamics.

Aspect level sentiment classification aims to classify the sentiment polarity of a given aspect term or aspect category in a sentence. For sentiment classification towards a given aspect term, since a sentence may contain more than one aspect term, there may exist some opinions which are not the modifiers of the given aspect term. It is necessary to capture relevant opinion for a certain aspect term. Previous works use the relative distance between an aspect term and all other words in a sentence, in order to capture the nearest opinion of the aspect term. This can be infeasible when the sentence has a complex syntactic structure. In this paper, we detect the dependency relation between an aspect term and its related sentiment words in the dependency parse tree. Then, we integrate this relationship into CNN and Bi-LSTM respectively. Experiments show that the related sentiment features for an aspect term is helpful for models to discriminate its sentiment polarity, and our proposed models achieve state-of-the-art results among neural network models.

In this paper, we introduce a new neonatal facial expression database for pain analysis. This database, called facial expression of neonatal pain (FENP), contains 11000 neonatal facial expression images associated with 106 Chinese neonates from two children's hospitals, i.e., the Children's Hospital Affiliated to Nanjing Medical University and Second Affiliated Hospital Affiliated to Nanjing Medical University in China. The facial expression images cover four categories of facial expressions, i.e., severe pain expression, mild pain expression, crying expression and calmness expression, where each category contains 2750 neonatal facial expression images. Based on this database, we also investigate the pain facial expression recognition problem using several state-of-the-art facial expression features and expression recognition methods, such as Gabor+SVM, LBP+SVM, HOG+SVM, LBP+HOG+SVM, and several Convolutional Neural Network (CNN) methods (including AlexNet, VGGNet, GoogLeNet, ResNet and DenseNet). The experimental results indicate that the proposed neonatal pain facial expression database would be very suitable for the study of both neonatal pain and facial expression recognition. Moreover, the FENP database would be publicly available after signing a license agreement (the users can contact Jingjie Yan (yanjingjie@njupt.edu.cn) or Guanming Lu (lugm@njupt.edu.cn)).

Current works on image emotion distribution learning typically extract visual representations from the holistic image or explore emotion-related regions in the image from a global-wise perspective. However, different regions of an image contribute differently to the arousal of each emotion. Existing works do not deeply explore corresponding emotion-aware regions of each emotion in the image, nor do they fully capture the relationship between each emotion-aware region and the emotion labels. In this paper, we propose a novel attention based emotion distribution learning method, which can explore the emotion-related regions of images from the perspective of each emotion category, and can conduct region relationship learning. Specifically, we introduce a semantic guided attention detection network to generate class-wise attention maps for each emotion and a global-wise attention map for the holistic image. Meanwhile, an emotional graph-based network is adopted to capture the correlation between each region and the emotion distribution. Experiments on several benchmark datasets demonstrate the superiority of the proposed method compared to related works.

The focal role of flow in promoting desirable outcomes in companies, such as increased employees' well-being and performance, led scholars to study flow in the context of work. However, current measurement approaches which assess flow via self-report scales after task execution are limited due to obtrusiveness and a lack of real-time support. Hence, new measurement approaches must be created to overcome these limitations. In this paper, we use cardiac features (heart rate variability; HRV) and a Random Forest classifier to distinguish high and low flow. Our results from a large-scale lab experiment with 158 participants and a field study with nine participants reveal, that with HRV features alone, flow-classifiers can be built with an accuracy of 68.5% (lab) and 70.6% (field). Our research contributes to the challenge of developing a less obtrusive, real-time measurement method of flow based on physiological features and to investigate flow from a physiological perspective. Our findings may serve as foundation for future work aiming to build physio-adaptive systems which can improve employee's performance. For instance, these systems could ensure that no notifications are forwarded to employees while the system is 'sensing' flow.

Deep learning methods have achieved significant progress in the presence of correctly annotated datasets in instance segmentation. However, object classes in large-scale datasets are sometimes ambiguous, which easily causes confusion. Besides, limited experience and knowledge of annotators can lead to mislabeled object semantic classes. To solve this issue, a novel method is proposed in this paper, which considers different roles of noisy class labels in different sub-tasks. Our method is based on two basic observations: firstly, the foreground-background annotation of a sample is correct even though its class label is noisy. Secondly, symmetric loss benefits the model robustness to noisy labels but harms the learning of hard samples, while cross entropy loss is the opposite. Based on the two basic observations, in the foreground-background sub-task, cross entropy loss is used to fully exploit correct gradient guidance. In the foreground-instance sub-task, symmetric loss is used to prevent incorrect gradient guidance provided by noisy class labels. Furthermore, we apply contrastive self-supervised loss to update features of all foreground, to compensate for insufficient guidance provided by partially correct labels especially in the highly noisy setting. Extensive experiments conducted with three popular datasets (i.e., Pascal VOC, Cityscapes and COCO) have demonstrated the effectiveness of our method in a wide range of noisy class label scenarios.

Although multiple techniques have been proposed with the goal of minimizing the semantic gap in virtual machine introspection, most concentrate on passive observation of the internal state, while there are also a number of proposals with which active modification of the VM's internal state is made possible. However there are issues when modifications are applied, such as keeping a consistent kernel state and avoiding a crash. In this paper we propose Oxpecker, a VMI platform for transactional modification. The out-of-VM read access allows an introspector to detect malware in the guest OS (e.g., rootkit) and the transactional write access allows Oxpecker to reliably neutralize the detected threats. To begin a transaction, Oxpecker monitors VM state changes waiting for an idle moment which is free of possible race-conditions in the guest kernel memory. Thereafter, it invokes a VMI client's callback to proceed with reading/writing in its memory. Upon user request or possible exceptions, transaction is rolled back while the transaction ACID properties are maintained at all times. Oxpecker is implemented and evaluated under different real-world workloads. Additionally and as a practical example, a tool is developed, and open sourced, based on Oxpecker with which guest VM processes could be killed.

As a rising star of social apps, short video apps, e.g., TikTok, have attracted a large number of mobile users by providing fresh and short video contents that highly match their watching preferences. Meanwhile, the booming growth of short video apps imposes new technical challenges on the existing computation and communication infrastructure. Traditional solutions maintain all videos on the cloud and stream them to users via contend delivery networks or the Internet. However, they incur huge network traffic and long delay that seriously affect users' watching experiences. In this paper, we propose an edge-assisted short video sharing framework to address these challenges by caching some videos highly preferred by users at edge servers that can be accessed by users via high-speed network connections. Since edge servers have limited computation and storage resources, we design an online algorithm with provable approximation ratio to decide which videos should be cached at edge servers, without the knowledge of future network quality and watching preferences changes. Furthermore, we improve the performance by jointly considering video fetching and user-edge association. Extensive simulations are conducted to evaluate the proposed algorithms under various system settings, and the results show that our proposals outperform existing schemes.

Network function virtualization achieves cost-effective network service provisioning through exploitation of virtualization and automation by decoupling network functions (software) from dedicated hardware. The software of the devices can then be hosted by low-cost computation devices rather than by more expensive dedicated devices. To obtain a specific network service, the traffic flow is steered to go through a specific order of network functions that are hosted by cloud computing, and this network function sequence is known as a service function chaining (SFC). To allocate computation resources for network functions and bandwidth resources between network functions in a physical network is the SFC embedding problem. In this paper, we consider the SFC embedding problem in multi-domain networks, where no domain information, like domain topology and network resource, is disclosed among domains. We propose a new optimization algorithm based on column generation method to solve this problem, which is distributedly computed in domains. To further improve the scalability, we provide two heuristic algorithms. We selected two networks one large (158 nodes) and one small (18 nodes) to numerically validate the proposed algorithms and demonstrate that the acceptance ratio obtained by the heuristic algorithms is close (within 5.6%) to that of the optimal algorithm.

Workflow scheduling in cloud environments has become a significant topic in both commercial and industrial applications. However, it is still an extraordinarily challenge to generate effective and economical scheduling schemes under the deadline constraint especially for the large scale workflow applications. To address the issue, this paper investigates the cloud workflow scheduling problem with the aim of minimizing the whole cost of workflow execution whereas maintaining its execution time under a predetermined deadline. A novel knowledge-based adaptive discrete water wave optimization (KADWWO) algorithm is developed based on the problem-specific knowledge of cloud workflow scheduling. In the proposed KADWWO, a discrete propagation operator is designed based on the idle time knowledge of hourly-based cost model to adaptively explore the huge search space. The adaptive refraction operator is employed to avoid stagnation and expand the available resource pool. Meanwhile, the dynamic grouping based breaking operator is designed to exploit the excellent block structure knowledge of task allocation scheme and corresponding resource to intensify the local region and accelerate convergence. Extensive simulation experiments on the well-known scientific workflow demonstrate that the KADWWO approach outperforms several recent state-of-the-art algorithms.

This paper presents V-Sight, a network monitoring framework for programmable virtual networks in clouds. Network virtualization based on software-defined networking (SDN-NV) in clouds makes it possible to realize programmable virtual networks; consequently, this technology offers many benefits to cloud services for tenants. However, to the best of our knowledge, network monitoring, which is a prerequisite for managing and optimizing virtual networks, has not been investigated in the context of SDN-NV systems. As the first framework for network monitoring in SDN-NV, we identify three challenges: non-isolated and inaccurate statistics, high monitoring delay, and excessive control channel consumption for gathering statistics. To address these challenges, V-Sight introduces three key mechanisms: 1) statistics virtualization for isolated statistics, 2) transmission disaggregation for reduced transmission delay, and 3) pCollector aggregation for efficient control channel consumption. The evaluation results reveal that V-Sight successfully provides accurate and isolated statistics while reducing the monitoring delay and control channel consumption in orders of magnitude. We also show that V-Sight can achieve a data plane throughput close to that of non-virtualized SDN.

Searchable encryption (SE) is a powerful technology that enables keyword-based search over encrypted data becomes possible. However, most SE schemes focus on exact keyword search which can not tolerate misspellings and typos. Existing fuzzy keyword search schemes only support fuzzy search within a limited similarity threshold d, the storage cost will grow exponentially or the precision of search results will greatly decrease as $d$ increases. Moreover, the current fuzzy keyword ranked search schemes consider only the keyword weight, and disregard the influence of keyword morphology similarity on the ranking. In this paper, we propose a scalable fuzzy keyword ranked search scheme over encrypted data under hybrid clouds architecture. We use the edit distance to measure the similarity of keywords and design an edit distance algorithm over the encrypted database, in which our scheme achieves fuzzy keyword search for any similarity threshold d with constant storage size and accurate search results. Furthermore, we design a two-factor ranking function combining keyword weight with keyword morphology similarity, which is utilized to rank the search results and enhance system usability. Extensive experiments are performed to demonstrate the trade-off of efficiency and security of the proposed scheme.

Cloud service providers typically provide different types of virtual machines (VMs) to cloud users with various requirements. Thanks to its effectiveness and fairness, auction has been widely applied in this heterogeneous resource allocation. Recently, several strategy-proof combinatorial cloud auction mechanisms have been proposed. However, they fail to protect the bid privacy of users from being inferred from the auction results. In this paper, we design a differentially private combinatorial cloud auction mechanism (DPCA) to address this privacy issue. Technically, we employ the exponential mechanism to compute a clearing unit price vector with a probability proportional to the corresponding revenue. We further improve the mechanism to reduce the running time while maintaining high revenues, by computing a single clearing unit price, or a subgroup of clearing unit prices at a time, resulting in the improved mechanisms DPCA-S and its generalized version DPCA-M, respectively. We theoretically prove that our mechanisms can guarantee differential privacy, approximate truthfulness and high revenue. Extensive experimental results demonstrate that DPCA can generate near-optimal revenues at the price of relatively high time complexity, while the improved mechanisms achieve a tunable trade-off between auction revenue and running time., 12 pages, 6 figures

With the development of wireless communications and the pervasiveness of location-aware mobile electronic devices, location-based services (LBS) which can provide a convenient lifestyle for people have attracted considerable interest recently. However, there still exists the privacy disclosure problem of LBS today. To solve this problem, in this paper, we present a novel privacy-preserving LBS search scheme in outsourced cloud. In the proposed LBS search scheme, the LBS providers data are first outsourced to the cloud server in an encrypted method. Then, a registered user constructs a query model to obtain accurate LBS query results without divulging his/her location information and query attribute to the LBS provider and the cloud server. Specifically, based on the designed matrix encryption technology, the LBS search scheme can achieve privacy preservation of users query and confidentiality of LBS data in the outsourced cloud server. Through security analysis, we show that our scheme can resist various known security threats. The experimental results further show that our LBS search scheme greatly reduces the communication overhead and provides convenient search experience to the users.

Machine learning (ML) models are increasingly trained with distributed workers possessing heterogeneous resources. In such scenarios, model training efficiency may be negatively affected by \emph{stragglers}---workers that run much slower than others. Efficient model training requires eliminating such stragglers, yet for modern ML workloads, existing load balancing strategies are inefficient and even infeasible. In this paper, we propose a novel strategy, called \emph{semi-dynamic load balancing}, to eliminate stragglers of distributed ML workloads. The key insight is that ML workers shall be load-balanced at \emph{iteration boundaries}, being non-intrusive to intra-iteration execution. Based on it we further develop LB-BSP, an integrated worker coordination mechanism that adapts workers' load to their instantaneous processing capabilities---by right-sizing the sample batches at the synchronization barriers. We have designed distinct load tuning algorithms for ML in CPU clusters, in GPU clusters as well as in federated learning setups, based on their respective characteristics. LB-BSP has been implemented as a Python module for ML frameworks like TensorFlow and PyTorch. Our EC2 deployment confirms that LB-BSP is practical, effective and light-weight, and is able to accelerating distributed training by up to $54\%$ .

Benefiting from the powerful computing and storage capabilities of cloud services, data sharing in the cloud has been permeated across various applications including social networks, e-health and crowdsourcing transportation system. Intuitively, outsourcing data to untrusted cloud commonly raises concerns about data privacy breaches. To combat this, one approach is exploiting Broadcast Based Searchable Encryption (BBSE) for secure data sharing. Nevertheless, the latest proposed BBSE is still defective in either security or efficiency. In this paper, we propose ESPD, an Efficient, Scalable and Privacy-preserving Data sharing framework over encrypted cloud dataset. Different from previous works, ESPD supports sharing target data to multiple users with distinct secret keys, and keeps a constant ciphertext length with the changes of the amount of system users. This feature significantly improves search efficiency and makes ESPD scalable in real-world scenarios. We show a formal analysis to prove the security of ESPD in terms of file privacy, keyword privacy and trapdoor privacy. Also, extensive experiments on real-world dataset are conducted to indicate the desirable performance of ESPD compared to other similar schemes.

Outsourcing medical data to healthcare cloud has become a popular trend. Since medical data of patients contain sensitive personal information, they should be encrypted before outsourcing. However, information retrieval methods based on plaintext cannot be directly applied to encrypted data. In this paper, we present a new cryptographic primitive named conjunctive keyword search with secure channel free and autonomous path delegation function (AP-SCF-PECKS), which can be applied in scenarios where patients want to search for and autonomous delegate their private medical information without revealing their private key. Particularly, the proposed solution allows patients to set up multi-hop delegation path with their preferences, and the delegated doctors in the path can search for and access the patient's private medical information with priority from high to low. Patients can ensure that authorized doctors are always trustworthy, and unauthorized users cannot obtain the private medical information of patients. Moreover, the scheme supports the conjunctive keyword search, secure channel free, and is secure against chosen keyword attack, chosen ciphertext attack, and keyword guessing attack. The security of proposed scheme has been formally proved in the standard model. Finally, the performance evaluations demonstrate that the overhead of proposed scheme are modest for healthcare cloud scenarios.

Attribute-based encryption (ABE) has been widely used to provide fine-grained access control to encrypted data in cloud computing. However, in practice, since the attributes are shared by multiple users in the ABE system, it is difficult to identify malicious users which intentionally but implicitly reveal their own ABE secret keys for some purpose. In recent years, a substantial amount of work focused on designing efficient tracing mechanisms to identify such misbehaving users on ABE. In this paper, we propose an efficient generic construction of black-box traceable ABE to identify the traitors. Our construction supports (public) adaptive black-box traceability and has short keys and short ciphertexts that are independent of the number of users. Technically, our construction relies on a generic transformation from any attribute-based inner-product functional encryption (ABIPFE) scheme to black-box traceable ABE system. Furthermore, to instantiate our generic construction from standard lattices, we propose the first lattice-based ABIPFE scheme, which is proven weakly selective secure in the standard model.

Applications hosted in commercial clouds are typically multi-tier and comprise multiple tightly coupled virtual machines (VMs). Service providers (SPs) cater to the users using VM instances with different configurations and pricing depending on the location of the data center (DC) hosting the VMs. However, selecting VMs to host multi-tier applications is challenging due to the trade-off between cost and quality of service (QoS) depending on the placement of VMs. This paper proposes a multi-cloud broker model called CoMCLOUD to select a sub-optimal VM coalition for multi-tier applications from an SP with minimum coalition pricing and maximum QoS. To strike a trade-off between the cost and QoS, we use an ant-colony-based optimization technique. The overall service selection game is modeled as a first-price sealed-bid auction aimed at maximizing the overall revenue of SPs. Further, as the hosted VMs often face demand spikes, we present a parallel migration strategy to migrate VMs with minimum disruption time. Detailed experiments show that our approach can improve the federation profit up to 23% at the expense of increased latency of approximately 15%, compared to the baselines.

A key factor in big data analytics and artificial intelligence is the collection of user data from a large population. However, the collection of user data comes at the price of privacy risks, not only for users but also for businesses who are vulnerable to internal and external data breaches. To address privacy issues, local differential privacy (LDP) has been proposed to enable an untrusted collector to obtain accurate statistical estimation on sensitive user data (e.g., location, health, and financial data) without actually accessing the true records. As key-value data is an extremely popular NoSQL data model, there are a few works in the literature that study LDP-based statistical estimation on key-value data. However, these works have some major limitations, including supporting small key space only, fixed key collection range, difficulty in choosing an appropriate padding length, and high communication cost. In this paper, we propose a two-phase mechanism PrivKVM* as an optimized and highly-complete solution to LDP-based key-value data collection and statistics estimation. We verify its correctness and effectiveness through rigorous theoretical analysis and extensive experimental results.

To reduce the cost of human and material resources and improve the collaborations among medical systems, research laboratories and insurance companies for healthcare researches and commercial activities, electronic medical records (EMRs) have been proposed to shift from paperwork to friendly shareable electronic records. To take advantage of EMRs efficiently and reduce the cost of local storage, EMRs are usually outsourced to the remote cloud for sharing medical data with authorized users. However, cloud service providers are untrustworthy. In this paper, we propose an efficient, secure, and flexible EMR sharing system by introducing a novel cryptosystem called dual-policy revocable attribute-based encryption and tamper resistance blockchain technology. Our proposed system enables EMRs to be shared at a fine-grained level and allows data users to detect any unauthorized manipulation. Moreover, the key generation center can revoke malicious users without affecting the honest users. We provide the formal security model as well as the concrete scheme with security analysis. The experimental simulation and experimental analysis of our proposed scheme demonstrate that our proposed system has superior performances to the most relevant solutions.

Federated learning allows a large number of resource-constrained clients to train a globally-shared model together without sharing local data. These clients usually have only a few classes (categories) of data for training, where the data distribution is non-iid (not independent identically distributed). In this paper, we put forward the concept of category privacy for the first time to indicate which classes of data a client has, which is an important but ignored privacy goal in the federated learning with non-iid data. Although secure aggregation protocols are designed for federated learning to protect the input privacy of clients, we perform the first systematic study on category inference attack and demonstrate that these protocols cannot fully protect category privacy. We design a differential selection strategy and two de-noising approaches to achieve the attack goal successfully. In our evaluation, we apply the attack to non-iid federated learning settings with various datasets. On MNIST, CIFAR-10, AG\_news, and DBPedia datasets, our attack achieves >90% accuracy measured in F1-score in most cases. We further consider a possible detection method and propose two strategies to make the attack more inconspicuous.

purposes is increasing at an astonishing rate. More and more people rely on online sites for purchasing songs, apparels, books, rented movies etc. The competition between the online sites forced the web site owners to provide personalized services to their customers. So the recommender systems came into existence. Recommender systems are active information filtering systems that attempt to present to the user, information items in which the user is interested in. The websites implement recommender system feature using collaborative filtering, content based or hybrid approaches. The recommender systems also suffer from issues like cold start, sparsity and over specialization. Cold start problem is that the recommenders cannot draw inferences for users or items for which it does not have sufficient information. This paper attempts to propose a solution to the cold start problem by combining association rules and clustering technique. Comparison is done between the performance of the recommender system when association rule technique is used and the performance when association rule and clustering is combined. The experiments with the implemented system proved that accuracy can be improved when association rules and clustering is combined. An accuracy improvement of 36% was achieved by using the combination technique over the association rule technique.

Given a publicly available pool of machine learning models constructed for various tasks, when a user plans to build a model for her own machine learning application, is it possible to build upon models in the pool such that the previous efforts on these existing models can be reused rather than starting from scratch? Here, a grand challenge is how to find models that are helpful for the current application, without accessing the raw training data for the models in the pool. In this paper, we present a two-phase framework. In the upload phase, when a model is uploading into the pool, we construct a reduced kernel mean embedding (RKME) as a specification for the model. Then in the deployment phase, the relatedness of the current task and pre-trained models will be measured based on the value of the RKME specification. Theoretical results and extensive experiments validate the effectiveness of our approach.

In this paper, we propose a Hierarchical Multimodal Variational Encoder-Decoder (HMMVED) to predict the popularity of micro-videos by comprehensively leveraging the user information and micro-video content in a hierarchical fashion. In particular, the multimodal variational encoder-decoder framework encodes the input modalities to a lower dimensional stochastic variable, from which the popularity of micro-videos can be decoded. Considering the important role of users social influence in social media for information dissemination, a user encoderdecoder is designed to learn the user Gaussian embedding from the user information, which is informative about the coarsegrained popularity for the micro-videos uploaded by the user. The learned user embedding then acts as the prior of the micro-video latent embeddings in the micro-video encoder-decoder, where the refined posterior embedding of the micro-video is obtained by taking the multimodal content into consideration. The finegrained popularity of each micro-video is generated from the posterior embedding of the micro-video. Based on the multimodal extension of variational information bottleneck theory, we show that the learned latent embeddings of micro-videos are maximally expressive about the popularity whilst maximally compressing the information from input modalities. Extensive experiments conducted on two real-world datasets demonstrate the effectiveness of the proposed method. Codes and datasets are available at: https://github.com/JennyXieJiayi/HMMVED.

Currently, with the rapid development of mobile Internet, micro-video has become a prevailing format of user-generated contents (UGCs) on various social media platforms. Several studies have been conducted towards to understanding high-level micro-video semantics, such as venue categorization, memorability analysis, and popularity prediction. However, these approaches supported tasks with only a single output, which exhibited limitations when attempting to use them to resolve tasks with multiple outputs, especially the multi-label micro-video classification. To tackle this problem, in this paper, we propose a dual multi-modal low-rank decomposition (DMLRD) method for multi-label micro-video classification tasks. To learn more comprehensive micro-video representations, we first learn the low-rank-regularized modality-specific and modality-shared components by considering the consistency and the complementarity among modalities simultaneously. Meanwhile, the less descriptive power of each modality aroused by inherent properties can be solved. To obtain unseen label representations, we next construct a sparsity-regularized multi-matrix normal estimation term to jointly encode the latent relationship structures among labels and dimensions. Experiments on a publicly available large-scale micro-video dataset demonstrate the effectiveness of our proposed model over the state-of-art methods.

Recently, we have observed an exponential increase of user-generated content (UGC) videos. The distinguished characteristic of UGC videos originates from the video production and delivery chain, as they are usually acquired and processed by non-professional users before uploading to the hosting platforms for sharing. As such, these videos usually undergo multiple distortion stages that may affect visual quality before ultimately being viewed. Inspired by the increasing consensus that the optimization of the video coding and processing shall be fully driven by the perceptual quality, in this paper, we propose to study the quality of the UGC videos from both objective and subjective perspectives. We first construct a UGC video quality assessment (VQA) database, aiming to provide useful guidance for the UGC video coding and processing in the hosting platform. The database contains source UGC videos uploaded to the platform and their transcoded versions that are ultimately enjoyed by end-users, along with their subjective scores. Furthermore, we develop an objective quality assessment algorithm that automatically evaluates the quality of the transcoded videos based on the corrupted reference, which is in accordance with the application scenarios of UGC video sharing in the hosting platforms. The information from the corrupted reference is well leveraged and the quality is predicted based on the inferred quality maps with deep neural networks (DNN). Experimental results show that the proposed method yields superior performance. Both subjective and objective evaluations of the UGC videos also shed lights on the design of perceptual UGC video coding.

Weakly supervised object detection (WSOD aims to train object detectors by using only image-level annotations. Many recent works on WSOD adopt multiple instance detection networks (MIDN, which usually generate a certain number of proposals and regard proposal classification as a latent model learning within image classification. However, these methods tend to detect salient object, salient object parts and clustered objects due to lack of instance-level annotations during training. Thus a core issue is how to guarantee that the network learn as many objects with precise bounding boxes as possible. In this paper, we address this issue by exploiting the potential of proposal scores during training. We propose an adaptive instance refinement (AIR framework with three novel designs, which can be integrated with MIDN into a single network. Specifically, adaptive instance mining attempts to discover all positive instances according to the score distribution of proposals and their spatial similarity. Adaptive score modulation dynamically adjusts proposal scores to make the network focus more on instances with different difficulties in different training iterations. Adaptive knowledge refinement distills important information from all previous stages by the weighted average of proposal scores. The experimental results on the PASCAL VOC 2007 and 2012 benchmarks and the MS COCO benchmark demonstrate that AIR significantly improves the performance of the original MIDN and achieves the state-of-the-art results.

Expert systems and neural networks are new tools for the control of fermentation processes. With expert systems the fermentation plant and the process itself is modelled via a generalized, qualitative system description based on the experience of human experts. On the other hand neural networks and interpolating associative memories can learn the process behaviour directly by process observation. The paper at hand reports, how both control techniques can be combined for purposes like process supervision, modelling and optimization of biological plants.

In the modern era, Cloud Platforms are the most needed port to maintain documents remotely with proper security norms. The concept of cloud environments is similar to the network channel. Still, the Cloud is considered the refined form of network, in which the data can easily be stored into the server without any range restrictions. The data maintained into the remote server needs a high-security feature, and the processing power of data should be high to retrieve the data back from the respective server. In the past, there were several security schemes available to protect the remote cloud server reasonably. However, the attack possibilities over the cloud platform remain; only all the researchers continuously work on this platform without any delay. This paper introduces a hybrid data security scheme called the Improved Attribute-Based Encryption Scheme (IABES). This IABES combines two powerful data security algorithms: Advanced Encryption Standard (AES) and Attribute-Based Encryption (ABE) algorithm. These two algorithms are combined to provide massive support to the proposed approach of data maintenance over the remote cloud server with high-end security norms. This hybrid data security algorithm assures the data cannot be attacked over the server by the attacker or intruder in any case because of its robustness. The essential generation process generates a credential for the users. It cannot be identified or visible to anyone as well as the generated certificates cannot be extracted even if the corresponding user forgets the credentials. The only way to get back the certification is resetting the credential. The obtained results prove the accuracy level of the proposed cypher security schemes compared with the regular cloud security management scheme, and the proposed algorithm essential generation process is unique. No one can guess or acquire it. Even the person may be the service provider or server administrator. For all, the proposed system assures data maintenance over the cloud platform with a high level of security and robustness in Quality of Service.

The global sensitivity analysis method used to quantify the influence of uncertain input variables on the variability in numerical model responses has already been applied to deterministic computer codes; deterministic means here that the same set of input variables gives always the same output value. This paper proposes a global sensitivity analysis methodology for stochastic computer codes, for which the result of each code run is itself random. The framework of the joint modeling of the mean and dispersion of heteroscedastic data is used. To deal with the complexity of computer experiment outputs, nonparametric joint models are discussed and a new Gaussian process-based joint model is proposed. The relevance of these models is analyzed based upon two case studies. Results show that the joint modeling approach yields accurate sensitivity index estimatiors even when heteroscedasticity is strong.

Now a day's internet is the most valuable source of learning, getting ideas, reviews for a product or a service. Everyday millions of reviews are generated in the internet about a product, person or a place. Because of their huge number and size it is very difficult to handle and understand such reviews. Sentiment analysis is such a research area which understands and extracts the opinion from the given review and the analysis process includes natural language processing (NLP), computational linguistics, text analytics and classifying the polarity of the opinion. In the field of sentiment analysis there are many algorithms exist to tackle NLP problems. Each algorithm is used by several applications. In this paper we have shown the taxonomy of various sentiment analysis methods. This paper also shows that Support vector machine (SVM) gives high accuracy compared to Naive bayes and maximum entropy methods.

The screening of data sets is essential to modern technology. The use of classical group testing to isolate objects that are individually positive has become the standard experimental procedure in many applied settings. Work is just beginning in applying group testing techniques to the identification of subsets of objects that are collectively positive. This paper addresses the development of probabilistic group testing methods that lead to the identification of positive combinations of objects with specific applications to data mining.