Biblio

Documents are a common method of storing infor-mation and one of the most conventional forms of expression of ideas. Cloud servers store a user's documents with thousands of other users in place of physical storage devices. Indexes corresponding to the documents are also stored at the cloud server to enable the users to retrieve documents of their interest. The index includes keywords, document identities in which the keywords appear, along with Term Frequency-Inverse Document Frequency (TF-IDF) values which reflect the keywords' relevance scores of the dataset. Currently, there are no efficient methods to delete keywords from millions of documents over cloud servers while avoiding any compromise to the user's privacy. Most of the existing approaches use algorithms that divide a bigger problem into sub-problems and then combine them like divide and conquer problems. These approaches don't focus entirely on fine-grained deletion. This work is focused on achieving fine-grained deletion of keywords by keeping the size of the TF-IDF matrix constant after processing the deletion query, which comprises of keywords to be deleted. The experimental results of the proposed approach confirm that the precision of ranked search still remains very high after deletion without recalculation of the TF-IDF matrix.

Populations move across regions in search of better living possibilities, better life outcomes or going away from problems that affected their lives in the previous region they lived in. In the United States of America, this problem has been happening over decades. Intelligent Conversational Text-based Agents, also called Chatbots, and Artificial Intelligence are increasingly present in our lives and over recent years, their presence has increased considerably, due to the usability cases and the familiarity they are wining constantly. Using NLP algorithms for law in accessible platforms allows scaling of users to access a certain level of law expert who could assist users in need. This paper describes the motivation and circumstances of this problem as well as the description of the development of an Intelligent Conversational Agent system that was used by immigrants in the USA so they could get answers to questions and get suggestions about better legal options they could have access to. This system has helped thousands of people, especially in California

The robustness of supply chain networks (SCNs) against sequential topology attacks is significant for maintaining firm relationships and activities. Although SCNs have experienced many emergencies demonstrating that mixed failures exacerbate the impact of cascading failures, existing studies of sequential attacks rarely consider the influence of mixed failure modes on cascading failures. In this paper, a reinforcement learning (RL)-based sequential attack strategy is applied to SCNs with cascading failures that consider mixed failure modes. To solve the large state space search problem in SCNs, a deep Q-network (DQN) optimization framework combining deep neural networks (DNNs) and RL is proposed to extract features of state space. Then, it is compared with the traditional random-based, degree-based, and load-based sequential attack strategies. Simulation results on Barabasi-Albert (BA), Erdos-Renyi (ER), and Watts-Strogatz (WS) networks show that the proposed RL-based sequential attack strategy outperforms three existing sequential attack strategies. It can trigger cascading failures with greater influence. This work provides insights for effectively reducing failure propagation and improving the robustness of SCNs.

It is proposed to address existing methodological issues in the educational process with the development of intellectual technologies and knowledge representation systems to improve the efficiency of higher education institutions. For this purpose, the structure of relational database is proposed, it will store the information about defended dissertations in the form of a set of attributes (heuristics), representing the mandatory qualification attributes of theses. An inference algorithm is proposed to process the information. This algorithm represents an artificial intelligence, its work is aimed at generating queries based on the applicant preferences. The result of the algorithm's work will be a set of choices, presented in ranked order. Given technologies will allow applicants to quickly become familiar with known scientific results and serve as a starting point for new research. The demand for co-researcher practice in solving the problem of updating the projective thinking methodology and managing the scientific research process has been justified. This article pays attention to the existing parallels between the concepts of technical and human sciences in the framework of their convergence. The concepts of being (economic good and economic utility) and the concepts of consciousness (humanitarian economic good and humanitarian economic utility) are used to form projective thinking. They form direct and inverse correspondences of technology and humanitarian practice in the techno-humanitarian mathematical space. It is proposed to place processed information from the language of context-free formal grammar dissertation abstracts in this space. The principle of data manipulation based on formal languages with context-free grammar allows to create new structures of subject areas in terms of applicants' preferences.It is believed that the success of applicants’ work depends directly on the cognitive training of applicants, which needs to be practiced psychologically. This practice is based on deepening the objectivity and adequacy qualities of obtaining information on the basis of heuristic methods. It requires increased attention and development of intelligence. The paper studies the use of heuristic methods by applicants to find new research directions leads to several promising results. These results can be perceived as potential options in future research. This contributes to an increase in the level of retention of higher education professionals.

Coverage-guided testing has shown to be an effective way to find bugs. If we model coverage-guided testing as a search problem (i.e., finding inputs that can cover more branches), then its efficiency mainly depends on two factors: (1) the accuracy of the searching algorithm and (2) the number of inputs that can be evaluated per unit time. Therefore, improving the search throughput has shown to be an effective way to improve the performance of coverage-guided testing.In this work, we present a novel design to improve the search throughput: by evaluating newly generated inputs with JIT-compiled path constraints. This approach allows us to significantly improve the single thread throughput as well as scaling to multiple cores. We also developed several optimization techniques to eliminate major bottlenecks during this process. Evaluation of our prototype JIGSAW shows that our approach can achieve three orders of magnitude higher search throughput than existing fuzzers and can scale to multiple cores. We also find that with such high throughput, a simple gradient-guided search heuristic can solve path constraints collected from a large set of real-world programs faster than SMT solvers with much more sophisticated search heuristics. Evaluation of end-to-end coverage-guided testing also shows that our JIGSAW-powered hybrid fuzzer can outperform state-of-the-art testing tools.

We present a scalable optimization algorithm and its parallel implementation for two-stage stochastic programming problems of large-scale, particularly the security constrained optimal power flow models routinely used in electrical power grid operations. Such problems can be prohibitively expensive to solve on industrial scale with the traditional methods or in serial. The algorithm decomposes the problem into first-stage and second-stage optimization subproblems which are then scheduled asynchronously for efficient evaluation in parallel. Asynchronous evaluations are crucial in achieving good balancing and parallel efficiency because the second-stage optimization subproblems have highly varying execution times. The algorithm employs simple local second-order approximations of the second-stage optimal value functions together with exact first- and second-order derivatives for the first-stage subproblems to accelerate convergence. To reduce the number of the evaluations of computationally expensive second-stage subproblems required by line search, we devised a flexible mechanism for controlling the step size that can be tuned to improve performance for individual class of problems. The algorithm is implemented in C++ using MPI non-blocking calls to overlap computations with communication and boost parallel efficiency. Numerical experiments of the algorithm are conducted on Summit and Lassen supercomputers at Oak Ridge and Lawrence Livermore National Laboratories and scaling results show good parallel efficiency.

Network function virtualization (NFV) simplies the coniguration and management of security services by migrating the network security functions from dedicated hardware devices to software middle-boxes that run on commodity servers. Under the paradigm of NFV, the service function chain (SFC) consisting of a series of ordered virtual network security functions is becoming a mainstream form to carry network security services. Allocating the underlying physical network resources to the demands of SFCs under given constraints over time is known as the SFC deployment problem. It is a crucial issue for infrastructure providers. However, SFC deployment is facing new challenges in trading off between pursuing the objective of a high revenue-to-cost ratio and making decisions in an online manner. In this paper, we investigate the use of reinforcement learning to guide online deployment decisions for SFC requests and propose a Policy network Assisted Monte Carlo Tree search approach named PACT to address the above challenge, aiming to maximize the average revenue-to-cost ratio. PACT combines the strengths of the policy network, which evaluates the placement potential of physical servers, and the Monte Carlo Tree Search, which is able to tackle problems with large state spaces. Extensive experimental results demonstrate that our PACT achieves the best performance and is superior to other algorithms by up to 30% and 23.8% on average revenue-to-cost ratio and acceptance rate, respectively.

Nowadays, with the continuous increase in oil prices and the worldwide shift towards clean energy, all-electric vehicles are booming. Thence, these vehicles need widespread charging systems operating securely and reliably. Consequently, these charging systems need the most robust cybersecurity measures and strong authentication mechanisms to protect its user. This paper presents a new security scheme leveraging human biometrics in terms of iris recognition to defend against multiple types of cyber-attacks such as fraudulent identities, man-in-the-middle attacks, or unauthorized access to electric vehicle charging stations. Fundamentally, the proposed scheme implements a security mechanism based on the inherently unique characteristics of human eye biometric. The objective of the proposed scheme is to enhance the security of electric vehicle charging stations by using a low-cost and efficient authentication using k-Nearest Neighbours (KNN), which is a lightweight encryption algorithm.We tested our system on high-quality images obtained from the standard IITD iris database to search over the encrypted database and authenticate a legitimate user. The results showed that our proposed technique had minimal communication and computation overhead, which is quite suitable for the resource-limited charging station devices. Furthermore, we proved that our scheme outperforms other existing techniques.

Trust is an important emerging area of study in human-robot cooperation. Many studies have begun to look at the issue of robot (agent) capability as a predictor of human trust in the robot. However, the assumption that agent capability is the sole predictor of human trust could underestimate the complexity of the problem. This study aims to investigate the effects of agent-strategy and agent-capability in a visual search task. Fourteen subjects were recruited to partake in a web-based grid search task. They were each paired with a series of autonomous agents to search an on-screen grid to find a number of outlier objects as quickly as possible. Both the human and agent searched the grid concurrently and the human was able to see the movement of the agent. Each trial, a different autonomous agent with its assigned capability, used one of three search strategies to assist their human counterpart. After each trial, the autonomous agent reported the number of outliers it found, and the human subject was asked to determine the total number of outliers in the area. Some autonomous agents reported only a fraction of the outliers they encountered, thus coding a varying level of agent capability. Human subjects then evaluated statements related to the behavior, reliability, and trust of the agent. The results showed increased measures of trust and reliability with increasing capability. Additionally, the most legible search strategies received the highest average ratings in a measure of familiarity. Remarkably, given no prior information about capabilities or strategies that they would see, subjects were able to determine consistent trustworthiness of the agent. Furthermore, both capability and strategy of the agent had statistically significant effects on the human’s trust in the agent.

Cloud-based enterprise search services (e.g., AWS Kendra) have been entrancing big data owners by offering convenient and real-time search solutions to them. However, the problem is that individuals and organizations possessing confidential big data are hesitant to embrace such services due to valid data privacy concerns. In addition, to offer an intelligent search, these services access the user’s search history that further jeopardizes his/her privacy. To overcome the privacy problem, the main idea of this research is to separate the intelligence aspect of the search from its pattern matching aspect. According to this idea, the search intelligence is provided by an on-premises edge tier and the shared cloud tier only serves as an exhaustive pattern matching search utility. We propose Smartness at Edge (SAED mechanism that offers intelligence in the form of semantic and personalized search at the edge tier while maintaining privacy of the search on the cloud tier. At the edge tier, SAED uses a knowledge-based lexical database to expand the query and cover its semantics. SAED personalizes the search via an RNN model that can learn the user’s interest. A word embedding model is used to retrieve documents based on their semantic relevance to the search query. SAED is generic and can be plugged into existing enterprise search systems and enable them to offer intelligent and privacy-preserving search without enforcing any change on them. Evaluation results on two enterprise search systems under real settings and verified by human users demonstrate that SAED can improve the relevancy of the retrieved results by on average ≈24% for plain-text and ≈75% for encrypted generic datasets.

Deep learning has made remarkable achievements in various domains. Active learning, which aims to reduce the budget for training a machine-learning model, is especially useful for the Deep learning tasks with the demand of a large number of labeled samples. Unfortunately, our empirical study finds that many of the active learning heuristics are not effective when applied to Deep learning models in batch settings. To tackle these limitations, we propose a density weighted diversity based query strategy (DWDS), which makes use of the geometry of the samples. Within a limited labeling budget, DWDS enhances model performance by querying labels for the new training samples with the maximum informativeness and representativeness. Furthermore, we propose a beam-search based method to obtain a good approximation to the optimum of such samples. Our experiments show that DWDS outperforms existing algorithms in Deep learning tasks.

In recent years, more and more medical institutions have been using electronic medical records (EMRs) to improve service efficiency and reduce storage cost. However, it is difficult for medical institutions with different management methods to share medical data. The medical data of patients is easy to be abused, and there are security risks of privacy data leakage. The above problems seriously impede the sharing of medical data. To solve these problems, we propose a blockchain-based medical data sharing scheme with attribute-based searchable encryption, named BMDS. In BMDS, encrypted EMRs are securely stored in the interplanetary file system (IPFS), while corresponding indexes and other information are stored in a medical consortium blockchain. The proposed BMDS has the features of tamper-proof, privacy preservation, verifiability and secure key management, and there is no single point of failure. The performance evaluation of computational overhead and security analysis show that the proposed BMDS has more comprehensive security features and practicability.

Cloud computing is a technology that provides users with a large storage space and an enormous computing power. For privacy purpose, the sensitive data should be encrypted before being outsourced to the cloud. To search over the outsourced data, searchable encryption (SE) schemes have been proposed in the literature. An SE scheme should perform searches over encrypted data without causing any sensitive information leakage. To this end, a few security constraints were elaborated to guarantee the security of the SE schemes, namely, the keyword privacy, the trapdoor unlinkability, and the access pattern. The latter is very hard to be respected and most approaches fail to guarantee the access pattern constraint when performing a search. This constraint consists in hiding from the server the search result returned to the user. The non respect of this constraint may cause sensitive information leakage as demonstrated in the literature. To fix this security lack, we propose a method that allows to securely request and receive the needed documents from the server after performing a search. The proposed method that we call the access pattern hiding (APH) technique allows to respect the access pattern constraint. An experimental study is conducted to validate the APH technique.

In order to solve the problem of difficult verification of query results in searchable encryption, we used the idea of Shamir-secret sharing, combined with game theory, to construct a randomly verifiable multi-cloud server searchable encryption scheme to achieve the correctness of the query results in the cloud storage environment verify. Firstly, we using the Shamir-secret sharing technology, the encrypted data is stored on each independent server to construct a multi-cloud server model to realize the secure distributed storage and efficient query of data. Secondly, combined with game theory, a game tree of query server and verification server is constructed to ensure honesty while being efficient, and solve the problem of difficulty in returning search results to verify under the multi-cloud server model. Finally, security analysis and experimental analysis show that this solution effectively protects data privacy while significantly reducing retrieval time.

Due to the wide deployment of deep learning applications in safety-critical systems, robust and secure execution of deep learning workloads is imperative. Adversarial examples, where the inputs are carefully designed to mislead the machine learning model is among the most challenging attacks to detect and defeat. The most dominant approach for defending against adversarial examples is to systematically create a network architecture that is sufficiently robust. Neural Architecture Search (NAS) has been heavily used as the de facto approach to design robust neural network models, by using the accuracy of detecting adversarial examples as a key metric of the neural network's robustness. While NAS has been proven effective in improving the robustness (and accuracy in general), the NAS-generated network models run noticeably slower on typical DNN accelerators than the hand-crafted networks, mainly because DNN accelerators are not optimized for robust NAS-generated models. In particular, the inherent multi-branch nature of NAS-generated networks causes unacceptable performance and energy overheads.To bridge the gap between the robustness and performance efficiency of deep learning applications, we need to rethink the design of AI accelerators to enable efficient execution of robust (auto-generated) neural networks. In this paper, we propose a novel hardware architecture, NASGuard, which enables efficient inference of robust NAS networks. NASGuard leverages a heuristic multi-branch mapping model to improve the efficiency of the underlying computing resources. Moreover, NASGuard addresses the load imbalance problem between the computation and memory-access tasks from multi-branch parallel computing. Finally, we propose a topology-aware performance prediction model for data prefetching, to fully exploit the temporal and spatial localities of robust NAS-generated architectures. We have implemented NASGuard with Verilog RTL. The evaluation results show that NASGuard achieves an average speedup of 1.74× over the baseline DNN accelerator.

In the current scenario we are facing the issue of real view which is object deal with image or in virtual world for such kind of difficulties the Augmented Reality has came into existence (AR). This paper deal with Augmented Reality Search (ARS). In this Augmented Reality Search (ARS) just user have to make the voice command and the Augmented Reality Search (ARS) will provide you real view of that object. Consider real world scenario where a student searched for NIT Bangalore then it will show the real view of that campus.

Most of existing audio fingerprinting systems have limitations to be used for high-specific audio retrieval at scale. In this work, we generate a low-dimensional representation from a short unit segment of audio, and couple this fingerprint with a fast maximum inner-product search. To this end, we present a contrastive learning framework that derives from the segment-level search objective. Each update in training uses a batch consisting of a set of pseudo labels, randomly selected original samples, and their augmented replicas. These replicas can simulate the degrading effects on original audio signals by applying small time offsets and various types of distortions, such as background noise and room/microphone impulse responses. In the segment-level search task, where the conventional audio fingerprinting systems used to fail, our system using 10x smaller storage has shown promising results. Our code and dataset are available at https://mimbres.github.io/neural-audio-fp/.

Semantic metadata is an important means to promote the integration of information and services and improve the level of search and discovery automation. Aiming at the problems that machine is difficult to handle service metadata description and lack of information metadata description in current SWIM information services, this paper analyzes the methods of metadata sematic empowerment and mainstream semantic metadata standards related to air traffic control system, constructs the SWIM information, and service sematic metadata model based on semantic expansion. The method of semantic metadata model mapping is given from two aspects of service and data, which can be used to improve the level of information sharing and intelligent processing.

With the location-based services (LBS) booming, the volume of spatial data inevitably explodes. In order to reduce local storage and computational overhead, users tend to outsource data and initiate queries to the cloud. However, sensitive data or queries may be compromised if cloud server has access to raw data and plaintext token. To cope with this problem, searchable encryption for geometric range is applied. Geometric range search has wide applications in many scenarios, especially the circular range search. In this paper, a practical and secure circular range search scheme (PSCS) is proposed to support searching for spatial data in a circular range. With our scheme, a semi-honest cloud server will return data for a given circular range correctly without uncovering index privacy or query privacy. We propose a polynomial split algorithm which can decompose the inner product calculation neatly. Then, we define the security of our PSCS formally and prove that it is secure under same-closeness-pattern chosen-plaintext attacks (CLS-CPA) in theory. In addition, we demonstrate the efficiency and accuracy through analysis and experiments compared with existing schemes.

The existence of adversarial samples makes the security of machine learning models in practical application questioned, especially the black-box adversarial attack, which is very close to the actual application scenario. Efficient search for black-box attack samples is helpful to train more robust models. We discuss the situation that the attacker can get nothing except the final predict label. As for this problem, the current state-of-the-art method is Boundary Attack(BA) and its variants, such as Biased Boundary Attack(BBA), however it still requires large number of queries and kills a lot of time. In this paper, we propose a novel method to solve these shortcomings. First, we improved the algorithm for generating initial adversarial samples with smaller L2 distance. Second, we innovatively combine a swarm intelligence algorithm - Particle Swarm Optimization(PSO) with Biased Boundary Attack and propose PSO-BBA method. Finally, we experiment on ImageNet dataset, and compared our algorithm with the baseline algorithm. The results show that:(1)our improved initial point selection algorithm effectively reduces the number of queries;(2)compared with the most advanced methods, our PSO-BBA method improves the convergence speed while ensuring the attack accuracy;(3)our method has a good effect on both targeted attack and untargeted attack.

Visible light communication (VLC) is a promising technique in the fifth and beyond wireless communication networks. In this paper, a secure multiple-input single-output VLC network is studied, where simultaneous lightwave information and power transfer (SLIPT) is exploited to support energy-limited devices taking into account a practical non-linear energy harvesting model. Specifically, the optimal beamforming design problems for minimizing transmit power and maximizing the minimum secrecy rate are studied under the imperfect channel state information (CSI). S-Procedure and a bisection search is applied to tackle challenging non-convex problems and to obtain efficient resource allocation algorithm. It is proved that optimal beamforming schemes can be obtained. It is found that there is a non-trivial trade-off between the average harvested power and the minimum secrecy rate. Moreover, we show that the quality of CSI has a significant impact on achievable performance.

Data sharing through the cloud is flourishing with the development of cloud computing technology. The new wave of technology will also give rise to new security challenges, particularly the data confidentiality in cloud-based sharing applications. Searchable encryption is considered as one of the most promising solutions for balancing data confidentiality and usability. However, most existing searchable encryption schemes cannot simultaneously satisfy requirements for both high search efficiency and strong security due to lack of some must-have properties, such as parallel search and forward security. To address this problem, we propose a variant searchable encryption with parallelism and forward privacy, namely the parallel and forward private searchable public-key encryption (PFP-SPE). PFP-SPE scheme achieves both the parallelism and forward privacy at the expense of slightly higher storage costs. PFP-SPE has similar search efficiency with that of some searchable symmetric encryption schemes but no key distribution problem. The security analysis and the performance evaluation on a real-world dataset demonstrate that the proposed scheme is suitable for practical application.

Social Internet of Things (SIoT) is an extension of Internet of Things (IoT) that converges with Social networking concepts to create Social networks of interconnected smart objects. This convergence allows the enrichment of the two paradigms, resulting into new ecosystems. While IoT follows two interaction paradigms, human-to-human (H2H) and thing-to-thing (T2T), SIoT adds on human-to-thing (H2T) interactions. SIoT enables smart “Social objects” that intelligently mimic the social behavior of human in the daily life. These social objects are equipped with social functionalities capable of discovering other social objects in the surroundings and establishing social relationships. They crawl through the social network of objects for the sake of searching for services and information of interest. The notion of trust and trustworthiness in social communities formed in SIoT is still new and in an early stage of investigation. In this paper, our contributions are threefold. First, we present the fundamentals of SIoT and trust concepts in SIoT, clarifying the similarities and differences between IoT and SIoT. Second, we categorize the trust management solutions proposed so far in the literature for SIoT over the last six years and provide a comprehensive review. We then perform a comparison of the state of the art trust management schemes devised for SIoT by performing comparative analysis in terms of trust management process. Third, we identify and discuss the challenges and requirements in the emerging new wave of SIoT, and also highlight the challenges in developing trust and evaluating trustworthiness among the interacting social objects.

Optimization plays an important role in many problems that expect the accurate output. Security of the data stored in remote servers purely based on secret key which is used for encryption and decryption purpose. Many secret key generation algorithms such as RSA, AES are available to generate the key. The key generated by such algorithms are need to be optimized to provide more security to your data from unauthorized users as well as from the third party auditors(TPA) who is going to verify our data for integrity purpose. In this paper a method to optimize the secret key by using cuckoo search algorithm (CSA) is proposed.

As a promising solution of spectrum shortage, spectrum sharing has received tremendous interests recently. However, under different sharing policies of different licensees, the shared spectrum is heterogeneous both temporally and spatially, and is usually uncertain due to the unpredictable activities of incumbent users. In this paper, considering the spectrum uncertainty, we propose a spectrum sharing based delay-tolerant traffic off-loading (SDTO) scheme. To capture the available heterogeneous shared bands, we adopt a mesh cognitive radio network and employ the multi-hop transmission mode. To statistically guarantee the end-to-end (E2E) session request under the uncertain spectrum supply, we formulate the SDTO scheme into a stochastic optimization problem, which is transformed into a mixed integer nonlinear programming (MINLP) problem. Then, a coarse-fine search based iterative heuristic algorithm is proposed to solve the MINLP problem. Simulation results demonstrate that the proposed SDTO scheme can well schedule the network resource with an E2E session guarantee.