Showing papers by "Zhejiang Gongshang University published in 2016"

Block-Row Sparse Multiview Multilabel Learning for Image Classification

Abstract: In image analysis, the images are often represented by multiple visual features (also known as multiview features), that aim to better interpret them for achieving remarkable performance of the learning. Since the processes of feature extraction on each view are separated, the multiple visual features of images may include overlap, noise, and redundancy. Thus, learning with all the derived views of the data could decrease the effectiveness. To address this, this paper simultaneously conducts a hierarchical feature selection and a multiview multilabel (MVML) learning for multiview image classification, via embedding a proposed a new block-row regularizer into the MVML framework. The block-row regularizer concatenating a Frobenius norm ( ${F}$ -norm) regularizer and an $\boldsymbol {\ell }_{\textbf {2,1}}$ -norm regularizer is designed to conduct a hierarchical feature selection, in which the ${F}$ -norm regularizer is used to conduct a high-level feature selection for selecting the informative views (i.e., discarding the uninformative views) and the $\boldsymbol {\ell }_{\textbf {2,1}}$ -norm regularizer is then used to conduct a low-level feature selection on the informative views. The rationale of the use of a block-row regularizer is to avoid the issue of the over-fitting (via the block-row regularizer), to remove redundant views and to preserve the natural group structures of data (via the ${F}$ -norm regularizer), and to remove noisy features (the $\boldsymbol {\ell }_{\textbf {2,1}}$ -norm regularizer), respectively. We further devise a computationally efficient algorithm to optimize the derived objective function and also theoretically prove the convergence of the proposed optimization method. Finally, the results on real image datasets show that the proposed method outperforms two baseline algorithms and three state-of-the-art algorithms in terms of classification performance.

A Threshold Anonymous Authentication Protocol for VANETs

Abstract: Vehicular ad hoc networks (VANETs) have recently received significant attention in improving traffic safety and efficiency. However, communication trust and user privacy still present practical concerns to the deployment of VANETs, as many existing authentication protocols for VANETs either suffer from the heavy workload of downloading the latest revocation list from a remote authority or cannot allow drivers on the road to decide the trustworthiness of a message when the authentication on messages is anonymous. In this paper, to cope with these challenging concerns, we propose a new authentication protocol for VANETs in a decentralized group model by using a new group signature scheme. With the assistance of the new group signature scheme, the proposed authentication protocol is featured with threshold authentication, efficient revocation, unforgeability, anonymity, and traceability. In addition, the assisting group signature scheme may also be of independent interest, as it is characterized by efficient traceability and message linkability at the same time. Extensive analyses indicate that our proposed threshold anonymous authentication protocol is secure, and the verification of messages among vehicles can be accelerated by using batch message processing techniques.

Joint Spectrum and Power Allocation for D2D Communications Underlaying Cellular Networks

Abstract: This paper addresses the joint spectrum sharing and power allocation problem for device-to-device (D2D) communications underlaying a cellular network (CN). In the context of orthogonal frequency-division multiple-access systems, with the uplink resources shared with D2D links, both centralized and decentralized methods are proposed. Assuming global channel state information (CSI), the resource allocation problem is first formulated as a nonconvex optimization problem, which is solved using convex approximation techniques. We prove that the approximation method converges to a suboptimal solution and is often very close to the global optimal solution. On the other hand, by exploiting the decentralized network structure with only local CSI at each node, the Stackelberg game model is then adopted to devise a distributed resource allocation scheme. In this game-theoretic model, the base station (BS), which is modeled as the leader, coordinates the interference from the D2D transmission to the cellular users (CUs) by pricing the interference. Subsequently, the D2D pairs, as followers, compete for the spectrum in a noncooperative fashion. Sufficient conditions for the existence of the Nash equilibrium (NE) and the uniqueness of the solution are presented, and an iterative algorithm is proposed to solve the problem. In addition, the signaling overhead is compared between the centralized and decentralized schemes. Finally, numerical results are presented to verify the proposed schemes. It is shown that the distributed scheme is effective for the resource allocation and could protect the CUs with limited signaling overhead.

Privacy-Preserving-Outsourced Association Rule Mining on Vertically Partitioned Databases

Abstract: Association rule mining and frequent itemset mining are two popular and widely studied data analysis techniques for a range of applications. In this paper, we focus on privacy-preserving mining on vertically partitioned databases. In such a scenario, data owners wish to learn the association rules or frequent itemsets from a collective data set and disclose as little information about their (sensitive) raw data as possible to other data owners and third parties. To ensure data privacy, we design an efficient homomorphic encryption scheme and a secure comparison scheme. We then propose a cloud-aided frequent itemset mining solution, which is used to build an association rule mining solution. Our solutions are designed for outsourced databases that allow multiple data owners to efficiently share their data securely without compromising on data privacy. Our solutions leak less information about the raw data than most existing solutions. In comparison to the only known solution achieving a similar privacy level as our proposed solutions, the performance of our proposed solutions is three to five orders of magnitude higher. Based on our experiment findings using different parameters and data sets, we demonstrate that the run time in each of our solutions is only one order higher than that in the best non-privacy-preserving data mining algorithms. Since both data and computing work are outsourced to the cloud servers, the resource consumption at the data owner end is very low.

LBT-Based Adaptive Channel Access for LTE-U Systems

Abstract: Driven by the demand for more radio spectrum resources, mobile operators are looking to exploit the unlicensed spectrum as a complement to the licensed spectrum. LTE-unlicensed (LTE-U), also referred to as licensed-assisted access by the third generation partnership project, is an extension of the LTE standard operating on the unlicensed spectrum. To realize LTE-U, its coexistence with Wi-Fi systems is the main challenge and must be addressed. In this paper, a listen-before-talk access mechanism featuring an adaptive distributed control function protocol is adopted for the small base stations (SBSs), whereby the backoff window size is adaptively adjusted according to the available licensed spectrum bandwidth and the Wi-Fi traffic load to satisfy the quality-of-service requirements of small cell users and minimize the collision probability of Wi-Fi users. Meanwhile, both licensed and unlicensed spectrum bands are jointly allocated to optimize spectrum efficiency. An admission control mechanism is further developed for the SBS to limit collision with Wi-Fi traffic. Extensive simulation results show that the proposed schemes achieve fair and harmonious coexistence between LTE-U small cells and the surrounding Wi-Fi service sets and substantially outperform baseline non-adaptive channel access mechanisms in the unlicensed spectrum.

Mining High Utility Patterns in One Phase without Generating Candidates

Abstract: Utility mining is a new development of data mining technology. Among utility mining problems, utility mining with the itemset share framework is a hard one as no anti-monotonicity property holds with the interestingness measure. Prior works on this problem all employ a two-phase, candidate generation approach with one exception that is however inefficient and not scalable with large databases. The two-phase approach suffers from scalability issue due to the huge number of candidates. This paper proposes a novel algorithm that finds high utility patterns in a single phase without generating candidates. The novelties lie in a high utility pattern growth approach, a lookahead strategy, and a linear data structure. Concretely, our pattern growth approach is to search a reverse set enumeration tree and to prune search space by utility upper bounding. We also look ahead to identify high utility patterns without enumeration by a closure property and a singleton property. Our linear data structure enables us to compute a tight bound for powerful pruning and to directly identify high utility patterns in an efficient and scalable way, which targets the root cause with prior algorithms. Extensive experiments on sparse and dense, synthetic and real world data suggest that our algorithm is up to 1 to 3 orders of magnitude more efficient and is more scalable than the state-of-the-art algorithms.

The interaction between internal R&D and different types of external knowledge sourcing: an empirical study of Chinese innovative firms

Abstract: This paper investigates to what extent internal RD horizontal connections; value chain partners, and technology service providers. We find that both internal R&D activities and external knowledge sourcing have a positive effect on firms' innovation performance. Strong internal R&D capabilities also increase the effect of sourcing from value chain partners and horizontal connections, but we do not find support for complementarity between internal R&D and collaborations with universities and research labs. These findings jointly suggest that the mixture of different types of external knowledge partners in combination with internal R&D capabilities is crucial in understanding the role of open innovation in emerging economies.

Political Connections and Industrial Pollution: Evidence Based on State Ownership and Environmental Levies in China

Abstract: We investigate how state involvement in the ownership of non-listed entrepreneurial firms affects pollution fees levied by national and provincial governments in China (environmental levies). While the national government sets minimum environmental standards, provincial governments can enact requirements that exceed these minimums, and they are largely responsible for enforcing even the national standards, so environmental levies can measure concessions that provinces make to encourage development and employment. Furthermore, state ownership is a good proxy for a firm’s political connections, which can influence the relationship between the firm and the environmental authorities. We find that firms with state ownership pay lower environmental levies, which indicates that concessions are made for political or economic purposes. However, these concessions are conditional on the level of development of the province offering them, with better developed provinces providing fewer concessions.

Improvement of a Privacy Authentication Scheme Based on Cloud for Medical Environment

Abstract: Medical systems allow patients to receive care at different hospitals. However, this entails considerable inconvenience through the need to transport patients and their medical records between hospitals. The development of Telecare Medicine Information Systems (TMIS) makes it easier for patients to seek medical treatment and to store and access medical records. However, medical data stored in TMIS is not encrypted, leaving patients' private data vulnerable to external leaks. In 2014, scholars proposed a new cloud-based medical information model and authentication scheme which would not only allow patients to remotely access medical services but also protects patient privacy. However, this scheme still fails to provide patient anonymity and message authentication. Furthermore, this scheme only stores patient medical data, without allowing patients to directly access medical advice. Therefore, we propose a new authentication scheme, which provides anonymity, unlinkability, and message authentication, and allows patients to directly and remotely consult with doctors. In addition, our proposed scheme is more efficient in terms of computation cost. The proposed system was implemented in Android system to demonstrate its workability.

Performances, kinetics and mechanisms of catalytic oxidative desulfurization from oils

Abstract: Ultra-deep desulfurization technologies are critical for cleaner oils and consequent better air quality. Earlier efforts in this field focused on specific catalysts and their catalytic efficiencies, while current interest has shifted to the differences between homogenous and heterogeneous catalysis systems applied in catalytic oxidation desulfurization (ODS) as well as their advantages and disadvantages. In this review, catalysts using various supports were described and their catalytic activities in total oxidation of sulfur compounds were evaluated and commented meanwhile, taking hydroperoxide as oxidant. Then, the effects of reaction parameters on catalyst activities and the kinetics and mechanisms that were used for ODS from oils were reviewed. Under the same conditions, heterogeneous catalysts performed better than homogeneous catalysts. Leaching of active components, existence of N-containing compounds and excessive reaction temperature would deactivate catalysts in ODS. Besides, power-law kinetics equations, Langmuir–Hinshelwood mechanism, and “nucleophilic attack” reaction mechanism” will provide in-depth analysis of desulfurization process and catalysts deactivation. Future research needs on ODS are proposed including the development of novel carrier materials, the optimization of acid sites distribution and the better understanding of deep reaction mechanisms.

“Oral” Tribological Study on the Astringency Sensation of Red Wines

Abstract: Astringency is a specific oral sensation dominated by a dryness and puckering feeling and is one of the main quality factors for red wines, teas as well as some fruit products. Various hypotheses have been proposed to explain the mechanisms of the astringency sensation; however which prevails is still unknown. Here we used a tribological system to determine if the astringency is predominantly a physical perception. Mixtures of whole human saliva and typical astringent compounds such as tannins and red wines were evaluated for their lubrication behavior. Friction coefficient was measured over a wide range of lubrication speed. By comparing friction coefficient with human sensory results of astringency, a positive correlation was found. In particular the highest correlation was observed (R2 = 0.93) at a sliding speed of 0.075 mm/s. Results from this work demonstrate that astringency of polyphenol-rich products is a sensory perception via a physical stimulus and can be quantified using tribology techniques. Practical Applications There is an increasing interest in sensory astringency, particularly its signaling process in oral cavity and the quantitative assessment of its intensity. This is due to the fact that astringency is an important quality characteristic which strongly influences consumer's acceptability for many polyphenol-rich foods. This research achieved the first experimental approach to sensory astringency determination on red wines by using a tribological technique. Results from this work positively confirm the hypothesis that astringency is a tactile sensation. The method established in this work provides a useful tool for efficient determination of astringent intensity both as industrial quality control and for the prediction of consumer's sensory perception.