ISSN 0253-2778

CN 34-1054/N

open
Free to Publish. Free to Read.

2019 Vol. 49, No. 7

  Previous Issue | Next Issue 
Display Method:
Unsupervised identification of Malay domain multiword expressions
WANG Lin, LIU Wuying
2019, 49(7): 517-523. doi: 10.3969/j.issn.0253-2778.2019.07.001
Article Views:244 Full article PDF
Abstract:
Multiword expression (MWE) is an optimal granularity of language reuse. However, no explicit formal boundaries between MWEs and other words cause a serious problem on automatic identification of MWEs for some non-common languages. We address the identification issue of Malay domain MWEs, and propose a natural-annotation-based unsupervised extraction and clustering algorithm.In the novel algorithm, we firstly use a binary classification for each space character to solve length-varying Malay MWEs extraction, secondly transfer natural document-level category annotations to MWE-level ones for Malay MWEs clustering, and finally distill out several domain datasets of MWEs after filtering general MWEs.The experimental results in the Malay dataset of 272 783 text documents show that our algorithm can extract MWEs precisely and dispatch them into domain lexicons efficiently.
A novel method for mining latent events based on scalable patterns
QIU Zhen, WANG Qiyuan, LIU Di, MENG Hongmin
2019, 49(7): 524-532. doi: 10.3969/j.issn.0253-2778.2019.07.002
Article Views:177 Full article PDF
Abstract:
Big data reflect the people's living habits, social and natural laws. Data stream, one of the most important forms of manifestation, has a wide range of applications. In the field of practical application of data stream, the waveband consisted of continuous data point can show the abundant semantics. Therefore, it’s significant to take the pattern (waveband) as the granularity and expressive form of data stream.
A new method node importance evaluation based on multi-domain topology characteristics in complex networks
LIU Yan, RAO Yuan
2019, 49(7): 533-543. doi: 10.3969/j.issn.0253-2778.2019.07.003
Article Views:181 Full article PDF
Abstract:
Many efforts have been made to evaluate node importance in complex networks. However, some traditional methods based on node position in networks do not take into consideration the influence derived from multiple domain topology features, which leads to the low evaluation precision about node importance. To solve this problem, based on a deep analysis of such traditional methods as mixed degree decomposition (MDD) algorithm, a new method, named cluster and neighbor mixed decomposition method(CNMD),is proposed, which combines the global and local features of the complex network topology structure, and adopts in kinds of three-degree influence principle to represent the local features of the node.Extensive experiments on ten kinds of network datasets in different field show that the average resolution, the lowest and the highest resolution of all experimental datasets are 98.73%, 92.44% and 99.99%, respectively,which is obviously better than traditional methods, like MDD, Eksd and MCDWE algorithms.Therefore, CNMD method is not only suitable for multi-scale undirected network topology, but also applicable for evaluating node importance under all circumstances.
Selective ensemble of online sequential adaption ELMS-based adaptive network intrusion detection
HE Jiezhou, LIU Jinping, ZHANG Wuxia, MA Tianyu, TANG Zhaohui, XU Pengfei
2019, 49(7): 544-554. doi: 10.3969/j.issn.0253-2778.2019.07.004
Article Views:203 Full article PDF
Abstract:
The popularity of the Internet and network equipment and the diversity of access methods have brought great about convenience as well as huge security challenges. The ways and means of network intrusion are becoming more diversified and faster. Traditional intrusion detection methods are unable to meet the security monitoring requirements of an increasingly complex network environment in terms of effectiveness, adaptability and real-time.This paper proposes a network intrusion detection method based on selective learning of the online sequential Adaption Extreme Learning Machines (OAELMs), termed SEoOAELM-NID. Firstly, an OAELM construction method with online incremental update function is proposed,which can automatically set the optimal number of hidden nodes. Bagging strategy is used to train several OAELM sub-learners with certain independence. Then, based on the Margin Distance Minimization (Margin Distance Minimization) guidelines, the OAELM sublearner is integrated into the gain measure, and ensembled by selecting a partial sublearner with high gain. To get a highly Selective Ensemble of OAELM high generalization ability. SEoOAELM-NID has the advantages of automatic optimal setting of hidden nodes and online sequential update of ELM sub-learners, so it can effectively adapt to the intrusion detection requirements of various complex network environments; and by selecting some optimal sub-learners for integration, the accuracy and effectiveness of the final detection results are guaranteed, and online application is used. The test results on the NSL-KDD data set show that SEoOAELM-NID can achieve higher detection rates and fast recognition speeds for known and unknown intrusion types than single learner and traditional ensemble learning-based network intrusion detection methods.
AnomayDetect：An online distance-based anomaly detection algorithm
HUO Wenjun, WANG Wei, LI Wen
2019, 49(7): 555-563. doi: 10.3969/j.issn.0253-2778.2019.07.005
Article Views:320 Full article PDF
Abstract:
Anomaly detection is a key challenge in data mining which has a wide range of applications in the field of the Internet, including network security, image recognition and intelligent operation. In particular, intelligent operation has made great progress in recent years. Existing anomaly detection algorithms have many problems, such as low accuracy and inability to update automatically. The problem of anomaly detection in the context of intelligent operation and a practical need for high-accuracy, online and universal anomaly detection algorithms is studied. Based on the existing algorithms, an online distance-based anomaly detection algorithm is identified. Through the experiments on Yahoo Web-scope S5 dataset it is shown that the algorithm can detect anomalies successfully. A comparative study of several anomaly detectors verifies the effectiveness of the proposed algorithm.
Research on an OAuth2.0-based unified authentication system in the smart campus environment
GAO Baozhong, DU Shouyan, LI Xinzhi, WANG Xinhua
2019, 49(7): 564-571. doi: 10.3969/j.issn.0253-2778.2019.07.006
Article Views:190 Full article PDF
Abstract:
Based on the construction of Shandong Normal University’s smart campus system, this paper summarizes the research methods of building the smart campus authentication system, which introduces the system role and authorization procedure of OAuth2.0-Based authentication and authorization technology, and analyzes the concrete implementation of the smart campus authentication platform. By conducting security experiments and theoretical analysis, the security and reliability of campus data acquisition has been improved.
Mixed linear matrix completion model based on auxiliary information
SONG Hui, YANG Ming
2019, 49(7): 572-578. doi: 10.3969/j.issn.0253-2778.2019.07.007
Article Views:142 Full article PDF
Abstract:
The matrix completion technology has been applied in many fields in recent years. Using existing auxiliary information to perform matrix completion to improve the accuracy of the completion has attracted attention. A matrix completion model is proposed, which mixes bilinear and unilateral linear relationships, considering the correlation between row information and column information and their respective characteristics, so that the mixed linear model can approximate the original matrix entries. At the same time, the convergence of using the ADMM algorithm to solve the convex optimization problem is proved, and makes two sets of experiments with synthetic datasets and real datasets, which proves that the proposed method is more effective compared with the existing model using auxiliary information, whose error under RMSE evaluation standard was reduced by more than 25% than other methods.
Performance analysis of relaying networks based on non-orthogonal multiple access
DENG Chao, ZHAO Xiaoya, LI Xingwang
2019, 49(7): 579-587. doi: 10.3969/j.issn.0253-2778.2019.07.008
Article Views:140 Full article PDF
Abstract:
The performance of non-orthogonal multiple access(NOMA) fixed gain amplify-and-forward (AF) relaying networks is investigated, and the different links are considered with different fading channels. In particular, both transceiver hardware impairments and channel estimation are considered. At the receiver, the destination node processes the received information using selection combining (SC) algorithm. To analyze the performance of the users, the analytical closed-form expressions for the outage probability of users’ symbols are derived. In addition, the approximate analysis of the outage probability reveals the insights of the parameters for both hardware impairments and channel estimation on the network performance. Simulation results indicate that outage probability is limited by the levels of distortion noise and channel estimation error.
Distributed adaptive control of nonlinear vehicular platoons
ZHONG Bin, WANG Xinghu, SHENG Jie
2019, 49(7): 588-594. doi: 10.3969/j.issn.0253-2778.2019.07.009
Article Views:187 Full article PDF
Abstract:
The distributed adaptive control problem of vehicular platoon is considered. The vehicle longitudinal dynamics is modeled by a second-order nonlinear model with uncertain parameters. And the directed graph is used to describe the relationship of information transmission between vehicles. By utilizing dynamic gain technique, a distributed adaptive controller independent of the global platoon information is proposed. Moreover, when the split/join maneuver happens, it is shown that the control objective can be reached without redesigning the distributed controller as long as the interaction graph contains a directed spanning tree with the leading vehicle as the root. The numerical simulation results demonstrate the effectiveness of the proposed control law.
Completely-competitive-equilibrium-based crowdsensing pricing mechanism
LI Meixuan, SUN Yue, HUANG He, XIN Yu, BU Xiaofei
2019, 49(7): 595-602. doi: 10.3969/j.issn.0253-2778.2019.07.010
Article Views:179 Full article PDF
Abstract:
Crowdsensing accomplishes extended general and complex social sensing tasks through allocating tasks to a large number of ordinary users (or workers), and has attracted extensive attention in recent years. How to motivate users to participate in sensing tasks is one of the most important issues in crowdsensing. However, the existing incentive mechanisms mainly focus on how to set prices to enable users to submit high-quality sensing data,ignoring the problem of blind quotes, which can easily lead to the imbalance of the number of users participating in the task execution, so that the platform cannot obtain the optimal revenue. To tackle this challenge, a completely-competitive-equilibriumcrowdsensing pricing mechanism is proposed. Firstly, the multi-player game between platform and users is abstracted as a two-person game between the platform and the market. Then the market type probability is introduced and the two-person incomplete information game is transformed into the two-person complete imperfect information game through Harsanyi transformation. Finally, through multiple rounds of repeated games on the platform, the platform′s price converged to completely competitive equilibrium. Theoretical analysis and experimental results show that the proposed incentive mechanism can achieve completely competitive equilibrium.

Copyright © Editorial Office of JUSTC, All Rights Reserved. 皖ICP备05002528号

Supported by: Beijing Renhe Information Technology Co. Ltd