Most Down Articles

    Published in last 1 year | In last 2 years| In last 3 years| All| Most Downloaded in Recent Month | Most Downloaded in Recent Year|

    Most Downloaded in Recent Year
    Please wait a minute...
    A text classification model based on BiLSTM and label embedding
    DONG Yan-ru, LIU Pei-yu, LIU Wen-feng, ZHAO Hong-yan
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2020, 55 (11): 78-86.   DOI: 10.6040/j.issn.1671-9352.1.2019.024
    Abstract82)      PDF(pc) (1668KB)(550)       Save
    A text classification model based on BiLSTM and label embedding for text classification was proposed. Firstly, our method introduced the BERT model to extract high-quality sentence features. Then, we used BiLSTM and attention mechanism to get the text representations that integrate important context information. Finally, labels and words learned in the joint space, and we used the compatibility score deriving from the label-word pairs to weight the labels and sentences representations, realizing the double label embeddings. The classifier classifies sentences according to the given label information. Experimental results on five general authoritative datasets show that our method effectively improves the text classification performance, and our model has better practicability.
    Reference | Related Articles | Metrics
    A sentiment analysis method based on a polarity lexicon
    ZHANG Cheng-gong 1, 2, LIU Pei-yu1, 2*, ZHU Zhen-fang1,2, FANG Ming1,2
    J4    2012, 47 (3): 47-50.  
    Abstract1217)      PDF(pc) (1026KB)(2883)       Save

    A polarity lexicon is the foundation of sentiment analysis and orientation analysis. An overall and effective polarity lexicon was constructed, including base lexicon, domain lexicon, network lexicon and modifier lexicon. The influence of modifiers to polarity words was studied. A sentiment analysis method based on a polarity lexicon was proposed, in which the modifiers and polarity words were combined into polarity phrases, and the phrase was used as the basic unit to compute the polarity of sentences and texts. Experimental results showed that the effect of orientation analysis using polarity lexicon raised in this paper was good.

    Related Articles | Metrics
    Comparative study on genetic structure of three color variants of the sea cucumber (Apostichopus japonicus) based on mitochondrial and ribosomal genes
    Zhan-sheng GUO,Zhen WANG,Xu-guang HOU,Hai-tao ZHANG
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2020, 55 (11): 1-7.   DOI: 10.6040/j.issn.1671-9352.0.2020.231
    Abstract85)   HTML15)    PDF(pc) (869KB)(323)       Save

    The sea cucumber Apostichopus japonicus is an important marine aquatic species in China. Selecting good breeding is a major issue for sea cucumber aquaculture, and body color is chosen as an important consideration. In order to explore the genetic structure and phylogenetic relationship of sea cucumber populations with different body color characteristics, 16S rRNA, COⅠ and 18S rRNA-ITS-28S rRNA sequences were amplified and sequenced. The results showed that the lengths of 16S rRNA, COⅠ and 18S rRNA-ITS-28S rRNA sequences were 812-830 bp, 877-915 bp and 1 536-1 572 bp, respectively. In the green, white and purple morphs of the sea cucumber, 4/5/5, 5/3/4, and 4/4/3 haplotypes were detected in 16S rRNA, COⅠ and 18S rRNA-ITS-28S rRNA sequences, respectively. The number of polymorphic loci in the COⅠ sequences was the highest, and that of 16S rRNA sequences was the smallest. The genetic distances of 16S rRNA, COⅠ and 18S rRNA-ITS-28S rRNA sequences among three color morphs were both small, ranging from 0-0.014 7, 0-0.021 2, and 0-0.010 3, respectively, and they did not reach the species differentiation level. The phylogenetic tree was constructed based on the COⅠ sequences, and Strongylocentrotus purpuratus was chose as the outgroup. The results showed that the white sea cucumbers were clustered into one branch, the purple and green sea cucumbers were cross-aggregated with each other and could not be distinguished.

    Table and Figures | Reference | Related Articles | Metrics
    Survey on application of data mining via differential privacy
    KANG Hai-yan, MA Yue-lei
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2017, 52 (3): 16-23.   DOI: 10.6040/j.issn.1671-9352.2.2016.053
    Abstract709)      PDF(pc) (932KB)(1394)       Save
    The latest results of differential privacy in data mining were surveyed. The basic concepts of differential privacy were introduced. It analyzes the differential privacys research in pattern mining, classification and cluster. It was focused that on the analysis of the principle of some important technology to achieve. And also it was made that comparative analysis of its strengths and weaknesses and algorithm complexity. Finally, the future research of difference privacy under the dynamic data publication and big data environments was discussed.
    Reference | Related Articles | Metrics
    Study and application of CAD technology in grade crossing vertical design
    WANG Gui-Jie, WANG Wen-De, TAO De-Li
    J4    2009, 44 (11): 79-82.  
    Abstract794)      PDF(pc) (1883KB)(1548)       Save

    he advantages and disadvantages are analysed on the basis of the grade crossing vertical design. Based on these design methods and computer aided design(CAD)technology,grid and contour line construction drawings are given by auto draft.Design results could be checked by the threedimensional Surface model based on the grids data and the threedimensional panoramic dynamic perspective picture.Finally, as certification, some road grade crossing engineering cases which are designed by these methods and CAD technology are presented.

    Related Articles | Metrics
    Academic paper recommendation based on heterogeneous network embedding
    Kan XU,Rui-xin LIU,Hong-fei LIN,Hai-feng LIU,Jiao-jiao FENG,Jia-ping LI,Yuan LIN,Bo XU
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2020, 55 (11): 35-45.   DOI: 10.6040/j.issn.1671-9352.1.2019.017
    Abstract142)   HTML13)    PDF(pc) (4302KB)(384)       Save

    This paper focuses on the extraction and utilization of information from heterogeneous information network recommendation, and proposes a heterogeneous network embedding-based academic paper recommendation method. The proposed method uses the random walk strategy guided by the meta-path to generate the sequence of nodes. Then, for each meta-path, the unique embedding representation of nodes is learned by maximizing its co-occurrence probability with the adjacent nodes in the sequence obtained from the given meta-path. In addition, in order to fuse the low-dimensional representation of nodes in differentmetapaths into the final output of heterogeneous information network embedding, the different fusion functions are designed to achieve this goal, and attention mechanism is introduced to recommendation system. This paper applies the heterogeneous information network based on meta paths to solve the problem that most recommendation methods based on heterogeneous information networks rely on path similarity, which can not fully mine the potential structural characteristics of users and projects. We use DBLP data set to validate the effectiveness of the proposed model, which achieved better performance compared with the traditional model on RMSE.

    Table and Figures | Reference | Related Articles | Metrics
    On torsion free class and cover class in category of comodules
    LI Yuan, YAO Hai-lou
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2020, 55 (8): 1-5.   DOI: 10.6040/j.issn.1671-9352.0.2019.869
    Abstract279)      PDF(pc) (488KB)(389)       Save
    For a coalgebra, the concept of(pre)covers for comodules is first introduced and some properties about them are given. Then, the concepts of maximal tilting comodules and cover comodules are given, and the existence of a bijection between the tilting torsion free classes and the maximal tilting comodules is proved. Finally, when the tilting torsion free class is a cover class, whose unique representation by cover comodules over a coalgebra is obtained.
    Reference | Related Articles | Metrics
    Input-output finite time stability for event-triggered control of switched singular systems
    FENG Na-na, WU Bao-wei
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2019, 54 (3): 75-84.   DOI: 10.6040/j.issn.1671-9352.0.2018.297
    Abstract376)      PDF(pc) (545KB)(444)       Save
    The input-output finite-time stability for event-triggered-based switched singular systems is investigated. Concept of input-output finite-time stability for switched singular systems is given, and an event-triggered condition is proposed. The dynamic output feedback controller based on event-triggered mechanism is designed, by applying the Lyapunov function technique and average dwell time approach, some sufficient conditions for input-output finite-time stability of the switched singular closed-loop systems are derived, furthermore, dynamic output feedback controller parameters are obtained. Lastly, a numerical example is employed to illustrate the validity of the theoretical results.
    Reference | Related Articles | Metrics
    Optimal birth rate control for competing populations dependent on scale structure
    ZHANG Ping, LUO Zhi-xue
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2020, 55 (11): 18-25.   DOI: 10.6040/j.issn.1671-9352.0.2020.320
    Abstract79)      PDF(pc) (421KB)(367)       Save
    A competitive population model with scale structure-dependent is established, and the optimal birth-control problem of the model is discussed. Firstly, the existence and uniqueness of the system solution are proved by using the characteristic line method and the Banach fixed point theorem, and the continuous dependence of the system solution on the control variables is also proved by the comparison principle. At the end of the paper, the necessity condition of the existence of optimal control is proved by using the definition of normal cone.
    Reference | Related Articles | Metrics
    Application impacts analysis on natural forestry management for pine plantation in central and southern mountainous area of Shandong province
    Chun-hui LING,Ming-zhong ZHANG,Gui-yun LYU,Hao-ran CUI,Pan YAP,Ye WEI,Sheng-yuan LIU,Feng-yun MA,Hua-tian WANG,Xiu-mei LIU,Ling MA
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2020, 55 (11): 8-17.   DOI: 10.6040/j.issn.1671-9352.0.2020.199
    Abstract74)   HTML12)    PDF(pc) (2192KB)(358)       Save

    The near-natural management experiments were carried out at three forestry stations of Culaishan, Taishan and Mingguangsi, with long-time thinning possesses of 2, 10 and 3 years, respectively. The results revealed that: (1) The effects of thinning management could promote the seedling growth compared to the controlled treatment (CK), the main height, diameter at breast height (DBH) and canopy diameter increased by 3.13%-12.89%, 3.03%-8.91% and 2.75%-25.91%, respectively. Similarly, the annual mean diameter of residual trees height, DBH and diameter of canopy also improved by 21.43%-66.67%, 9.21%-26.15% and 64.29%-300% when suffered to near-natural processing. (2) The intermediate cutting changed the canopy gap, led to a decrease in crown closure. On the contrary, it caused an increase in plant species of shrub layer especially for the light-loving plants like Platycladus orientalis and Ailanthus altissima. Consistently, it could contribute to the herbs diversity, which caused improvements of diversity indexes like Simpson, Shannon-wiener and Gleason in pure pine stand and mixed pine-oak forest compared with that in CK. At the same time, the important values of Digitaria sanguinalis, Spodiopogon sibiricus and Carex rigescens elevated. (3) The tending measurement resulted in an improvement of soil pH, and a decrease of bulk density. As well, it had a potential to accelerate the nutrient metabolism for the increasing enzyme activities of catalase, phosphatase, sucrase and urease, which increased by 6.74%-29.33%, 4.79%-4.99%, 2.13%-10.34% and 1.25%-25.56%, respectively. It could be seen that the tending operation could benefit to the stand growth, enrich the undergrowth diversity, also improve the soil quality in different stand conditions.

    Table and Figures | Reference | Related Articles | Metrics
    Class-specific β distribution reduction in interval-valued decision systems
    HAN Shuang-zhi, ZHANG Nan, ZHANG Zhong-xi
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2020, 55 (11): 66-77.   DOI: 10.6040/j.issn.1671-9352.0.2020.268
    Abstract80)      PDF(pc) (2678KB)(357)       Save
    Attribute reduction is one of the important research points in rough set theory. The goal of β distribution reduction in interval-valued decision systems is to keep the corresponding β distribution of objects unchanged. In actual needs, attribute reduction usually only needs to focus on specific decision class rather than all decision classes. This paper proposes a theoretical of class-specific β distribution reduction in interval-valued decision systems. First of all, the basic concept of class-specific β distribution reduction is defined, and then the discernibility matrix corresponding to the class-specific β distribution is constructed. Finally, a class-specific β distribution reduction algorithm based on the discernibility matrices is proposed. In experiments, six UCI data sets are used to compare reduction results and reduction efficiency of BRADM algorithm and CSBRADM algorithm. The experiments results show that the reduction results of class-specific algorithm can keep the β distribution for class-specific unchanged, and reduction length of algorithm for specific class is less than or equal to reduction length of algorithm for all classes, and the CSBRADM algorithm efficiency is higher than the BRADM algorithm.
    Reference | Related Articles | Metrics
    Emergency decisions and coordination of closed-loop supply chain with production diseconomies under cost disruptions
    ZHONG Wei-shuo, DAI Geng-xin, MU Zong-yu
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2020, 55 (11): 96-110.   DOI: 10.6040/j.issn.1671-9352.0.2020.008
    Abstract78)      PDF(pc) (612KB)(331)       Save
    Aiming at the production diseconomies scale of manufacturer, the decentralized decision-making system and centralized decision-making system of manufacturer recycling and retailer recycling are constructed respectively. The emergency management problem of closed-loop supply chain under production cost disturbance is studied. The results show that: when the production cost of products is small, all kinds of equilibrium decisions made by decentralized decision-making system and centralized decision-making system under stable operation environment are robustness, and with the increase of the scale diseconomies elasticity coefficient, the disturbance interval with robustness of the optimal equilibrium decision-making in the stable operation environment is also increasing. When the production cost of the product is disturbed greatly, the retail price should be adjusted according to the disturbance direction, the product demand should be adjusted in the opposite direction, and the recycling price of waste products should be remain unchanged. In the selection of emergency recovery mode, it is better for manufacturers to directly recycle waste products. The designed contract can realize the coordination of the closed loop supply chain under normal operating environment and after disturbance.
    Reference | Related Articles | Metrics
    Fault tracing method of industrial production control data network based on SDG simplified model
    Yan-hua YANG,Li-gang YAO
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2020, 55 (11): 46-57.   DOI: 10.6040/j.issn.1671-9352.0.2020.353
    Abstract85)   HTML9)    PDF(pc) (1763KB)(314)       Save

    For the fault diagnosis and source tracing in industrial production control data network, this paper proposes a solution based on SDG simplified model. Firstly, a SDG model is established based on the structural characteristics of the industrial production control data network, then the search space for fault location is reduced through hierarchical and demarcated modeling method.On this basis, according to the observed signal amount extracted when the business alarm occurs, analyse the survival of the nodes based on the longitudinal data and survival data. Secondly, combined with the analysis of the characteristics of network faults, the single-node fault detemination method is used to determine whether each node device is faulty. Finally, based on the fault cause classification of the fault tracing database in the traceability diagnosis rules, the root cause of the network fault is determined. Through the test and verification method, the feasibility of the proposed method in the industrial production control data network is verified, which can meet the requirements of rapid fault tracing.

    Table and Figures | Reference | Related Articles | Metrics
    A vTPM-VM live migration scheme based on KVM
    HUANG Yu-qing, ZHAO Bo, XIAO Yu, TAO Wei
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2017, 52 (6): 69-75.   DOI: 10.6040/j.issn.1671-9352.2.2016.231
    Abstract718)      PDF(pc) (1919KB)(813)       Save
    The virtual machine equipped with virtual Trusted Platform Module(vTPM)could not live migrate in KVM platform. To solve this problem, a live migration scheme of virtual machine equipped vTPM based on KVM is proposed. It is analyzed that the KVM architecture and virtualization features of vTPM and integrated the live migration of virtual machines equipped vTPM with those native normal virtual machines in KVM, thus to ensure the consistency of security state in system before and after the migration and the safety of vTPM instance data during the migration process. Finally, the scheme and did the experiment is realized. The results showed that compared with the live migration of normal virtual machine, our method make the user be unaware of the migration process. Meanwhile, the average downtime of virtual machine equipped with vTPM during the migration process is no more than 50 ms and the performance loss is 15% and after the migration, the user could use the vTPM functions properly.
    Reference | Related Articles | Metrics
    BERT-IDCNN-CRF for named entity recognition in Chinese
    Ni LI,Huan-mei GUAN,Piao YANG,Wen-yong DONG
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2020, 55 (1): 102-109.   DOI: 10.6040/j.issn.1671-9352.2.2019.076
    Abstract1519)   HTML42)    PDF(pc) (2001KB)(356)       Save

    The pre-trained language model, BERT (bidirectional encoder representations from transformers), has shown promising result in NER (named entity recognition) due to its ability to represent rich syntactic, grammatical information in sentences and the polysemy of words. However, most existing BERT fine-tuning based models need to update lots of model parameters, facing with expensive time cost at both training and testing phases. To handle this problem, this work presents a novel BERT based language model for Chinese NER, named BERT-IDCNN-CRF (BERT-iterated dilated convolutional neural network-conditional random field). The proposed model utilizes traditional BERT model to obtain the context representation of the word as the input of IDCNN-CRF. At training phase, the model parameters of BERT in the proposed model remain unchanged so that the proposed model can reduce parameters training while maintaining polysemy of words. Experimental results show that the proposed model obtains significant training time with acceptable test error.

    Table and Figures | Reference | Related Articles | Metrics
    Cooperative group signature scheme with controllable linkability
    LIU Xin, XU Qiu-liang, ZHANG Bo
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2016, 51 (9): 18-35.   DOI: 10.6040/j.issn.1671-9352.0.2016.003
    Abstract466)      PDF(pc) (2837KB)(615)       Save
    The existing group signature schemes with controllable linkability(abbreviated as GS-OL)suffer from the following two deficiencies, i.e., group members’ private keys are likely to leak and the Open Authority may harm users’ privacy. Moreover, there is still room for improvement in the computational efficiency of users. To correct this situation, an extension of the existing GS-OL security model was made, and an improved scheme with higher efficiency was provided. By using the Boneh-Boyen signature scheme without pairings, the users were liberated from performing expensive pairing computations. Borrowing the ideas of cooperative group signatures, the problem of private key leakage of group members was also solved. Moreover, in order to decentralize the Open Authoritys power, the technique of distributed key generation, committed zero-knowledge proof of knowledge, and distributed proof of knowledge of preimages of group homomorphism were incorporated in the new scheme. In addition, a practical anonymous subscription system based on the new GS-OL scheme was proposed. Compared with previous systems, the new system meets several practical properties such as allowing customers’ tokens to expire and supporting privacy-preserving data mining, and it has obvious advantage in users’ operational efficiency.
    Reference | Related Articles | Metrics
    Security analysis of the eCK model
    NI Liang1,2,3, CHEN Gong-liang3, LI Jian-hua3
    J4    2013, 48 (7): 46-50.  
    Abstract621)      PDF(pc) (929KB)(1094)       Save

    Authenticated key agreement (AKA) protocol should capture desirable security properties as many as possible. As a formal method recently proposed to design and analyze two-party AKA protocols, the eCK model is currently receiving more and more attention, and the exact security guarantees that can be provided by this model are worthy of in-depth research. Hence, the relationship between the eCK model and the basic desirable security properties for AKA protocols is analyzed in detail. The conclusions indicate that AKA protocols with provable security in the eCK model capture most basic desirable security properties. Thereafter, the advantages and disadvantages of the eCK model are also summarized.

    Related Articles | Metrics
    Asymptotics for the LASSO estimator for partially linear models
    LI Feng1, LU Yi-qiang2
    J4    2012, 47 (3): 93-97.  
    Abstract760)      PDF(pc) (908KB)(1803)       Save

     Based on the profile least squares method, the LASSO penalty profile least squares estimator is constructed, and the choices of penalty parameter and bandwidth are also discussed. Because the optimization problem is linear, it can be easily implemented. Under some regular conditions, the consistency and asymptotic normality of the estimator for parameter component are investigated. Finally, Monte Carlo simulation studies are conducted to assess the finite sample performance of the proposed variable selection procedures.

    Related Articles | Metrics
    Optimality conditions on nonsmooth vector interval-valued optimization
    QIAN Xin-qiang, WANG Kai-rong
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2020, 55 (11): 26-34.   DOI: 10.6040/j.issn.1671-9352.0.2019.141
    Abstract84)      PDF(pc) (390KB)(292)       Save
    By using Clarke directional derivative and Clarke subdifferential, Fritz John optimal necessary conditions for weak LU efficient solutions of nonsmooth vector interval-valued optimization are obtained. Under the assumption of generalized invariant convexity and regularity of functions, KKT necessary optimality conditions, sufficient optimality conditions and related duality results are given. Some examples are used to verify the feasibility of the theory. These conclusions can solve the related problems of vector interval-valued optimization in general.
    Reference | Related Articles | Metrics
    Extremal eigenvalues of a class of tridiagonal interval matrices
    JIAN Yuan, LIU Ding-you
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2015, 50 (12): 15-22.   DOI: 10.6040/j.issn.1671-9352.0.2014.549
    Abstract598)      PDF(pc) (619KB)(690)       Save
    A class of tridiagonal interval matrices is studied by the recursive characteristic polynomials and Chebyshev polynomials where some results on eigenvalues are involved. The sharp (lower and upper) bound for the eigenvalues of this class of tridiagonal interval matrices is presended, and the realization matrices whose smallest (largest) eigenvalues reach the lower(upper) bound is characterized.
    Reference | Related Articles | Metrics
    Discrete sine cosine algorithm for solving large-scale 0-1 knapsack problems
    ZHENG Jian
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2020, 55 (11): 87-95.   DOI: 10.6040/j.issn.1671-9352.4.2020.109
    Abstract79)      PDF(pc) (552KB)(282)       Save
    According to the mathematical characteristics of the 0-1 knapsack problem(0-1 KP), this paper redesigns a discrete version of SCA(DSCA)for 0-1 KP. Within the framework of basic SCA, DSCA uses the real code to generate initial individuals, a new nonlinear exponential decreasing function is applied to adjust the individual update step size. A greedy-based repair operator is included to fix and optimize the infeasible solution. The performance of the improved algorithm was tested on two sets of large-scale 0-1 KP. The comparison with some state-of-arts algorithms confirms that DSCA is efficient and concise, not only can it provide an effective solution for 0-1 KP, but also it expands the application fields of SCA.
    Reference | Related Articles | Metrics
    Pythagorean fuzzy three-way concept lattice
    JI Ru-ya, WEI Ling, REN Rui-si, ZHAO Si-yu
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2020, 55 (11): 58-65.   DOI: 10.6040/j.issn.1671-9352.4.2020.241
    Abstract101)      PDF(pc) (543KB)(278)       Save
    Pythagorean fuzzy set theory is introduced into fuzzy three-way concept lattice. The construction of Pythagorean fuzzy three-way concept lattice is studied under the Pythagorean fuzzy formal context. First, the relationships between objects and attributes are expressed by the membership degree and non-membership degree combined with the Pythagorean fuzzy set theory. On this foundation, the definition of the Pythagorean fuzzy formal context is given. Next, based on threshold α, β and the idea of three-way decision, the object sets(the attribute sets)are divided into three parts: positive region, negative region and boundary region. On this basis, the definitions and relevant theorems of two kinds of Pythagorean fuzzy three-way concepts(the object induced Pythagorean fuzzy three-way concept and the attribute induced Pythagorean fuzzy three-way concept)are given, and the corresponding Pythagorean fuzzy three-way concept lattices are constructed. Finally, the applications of Pythagorean fuzzy three-way concept lattice in real problems are explained in detail with examples.
    Reference | Related Articles | Metrics
    Measuring user influence of a microblog based on information diffusion
    GUO Hao, LU Yu-liang, WANG Yu, ZHANG Liang
    J4    2012, 47 (5): 78-83.  
    Abstract621)      PDF(pc) (832KB)(1847)       Save

    Information diffusion and influence modeling are hot topics in microblog research. To do research on influence quantitatively, a concept based on the message diffusion was introduced and complied with how to count it out. The proposed approach was validated on real world datasets, and the result of experiments shows that our method is both effective and stable, especially in condition of limited dataset and time span.

    Related Articles | Metrics
    The preparation and purification of Supervillin antibody
    ZHANG Tong-tong, QIAN Ji-cheng, ZHU Chang-jun, DONG Zhi-xiong*
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2014, 49 (1): 25-30.   DOI: 10.6040/j.issn.1671-9352.0.2013.445
    Abstract434)      PDF(pc) (1589KB)(749)       Save

    Gelling spot protein Supervillin(SVIL) is a membrane protein molecule located in cells and extracellular matrix interface adhesive spots. To further study the cellular function of SVIL protein,we constructed plasmids of pGEXKG-SVILC302 and pHIS8-SVILC302 to express recombinant protein GST-SVILC302 and HIS-SVILC302 in E.coli BL21-DE3, respectively. The GST-SVILC302 protein was purified by Glutathione affinity chromatography and used to immune New Zealand white rabbits to generate the polyclonal antibody of SVIL. The HIS-SVILC302 protein was crosslinked with Ni-TNA Beads for affinity purification of the antibody. The results of Western Blot and Immunoflurescence experiments demonstrated that the antibody can specific recognize exogenous GFP-SVIL and endogenous SVIL protein. Our results showed that anti-SVIL polyclonal antibody we generated has high specificity and sensitivity enough to study the cellular functions of SVIL protein.

    Related Articles | Metrics
    A key management mechanism basedon TrustZone architecture
    CUI Xiao-yu, ZHAO Bo, FAN Pei-ru, XIAO Yu
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2017, 52 (6): 92-98.   DOI: 10.6040/j.issn.1671-9352.2.2016.219
    Abstract563)      PDF(pc) (1664KB)(561)       Save
    Aiming at the lack of encryption protection for user privacy data on embedded devices, a key management mechanism is proposed in TrustZone architecture. Combined with Physical Unclonable Function(PUF)technology and the users PIN to generate the user key related to both device and user, the confidentiality and integrity of privacy data is protected. Meanwhile, a relatively thorough key management mechanism is designed formutli-user data protection scene in public embedded device to realize key generation, storage, update and destroy. Finally, this multi-user key management moduleisgiven and realizedin TrustZone. And its security and efficiencyis analyzed by experiments. The experiment result show that the mechanism is reliable secure and has high key management efficiency.
    Reference | Related Articles | Metrics
    Short text classification based on word embedding similarity
    MA Cheng-long, JIANG Ya-song, LI Yan-ling, ZHANG Yan, YAN Yong-hong
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2014, 49 (12): 18-22.   DOI: 10.6040/j.issn.1671-9352.3.2014.295
    Abstract982)      PDF(pc) (2161KB)(2291)       Save
    As the short length of the Web short text and less shared words, a lot of out of vocabulary (OOV) words would appear, and these words make the task of text classification more difficult. To solve this problem, a new general framework based on word embedding similarity was proposed. First, get the word embedding file with unsupervised learning method based on unlabeled data. Second, extend the OOVs with the similar words in training data through computing the similarities of different word embeddings. The comparison with the baseline system shows that the proposed method gets better 1%-2% rate and outperforms more 10% rate on small training data set.
    Reference | Related Articles | Metrics
    Cut sets of outer P-fuzzy sets and extended rough sets models
    HAO Xiu-mei, LIU Ji-qin
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2020, 55 (10): 1-6.   DOI: 10.6040/j.issn.1671-9352.0.2020.196
    Abstract158)      PDF(pc) (384KB)(256)       Save
    The P-fuzzy sets are a pair of inner and outer P-fuzzy sets. Inner and outer P-fuzzy sets are two different kinds of dynamic fuzzy sets. The cut sets method is frequently used in fuzzy sets discussed. The concepts of λ-cut sets, strong λ-cut sets, interval cut sets and cut sets granularity are proposed, and the cut sets and interval cut sets granularity theorems of outer P-fuzzy sets are discussed. Then, the interval cut sets decomposition theorems of outer P-fuzzy sets are given. By using the concept of the outer P-fuzzy rough membership function, four rough sets extended models are proposed such as FF) rough sets, (1F,0F) rough sets, FF)probability rough sets and FF) variable precision rough sets. The FF)rough sets theorems of the outer P-fuzzy sets are discussed. In the end, the quantitative characteristics and relationship theorem are given.
    Reference | Related Articles | Metrics
    Piecewise linear Tobit regression model estimation based on kernel function method
    WANG Xiao-gang, LI Bing
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2020, 55 (6): 1-9.   DOI: 10.6040/j.issn.1671-9352.0.2019.865
    Abstract252)      PDF(pc) (609KB)(155)       Save
    In the change point estimation model, the common grid search method has the disadvantages of cumbersome calculation and unreal meaning, and the linearization technique cannot prove the large sample properties. In order to overcome these shortcomings, the kernel function method is proposed in the piecewise linear Tobit regression model to solve the problem of non-differentiable of the objective function, the change point and parameter estimation are obtained, the asymptotic properties are proved, and the validity and robustness of the estimation are verified by Monte Carlo simulation. The empirical analysis of the family annual financial asset data shows that family financial assets would be positively affected by the level of education, and the educational level would change before and after the undergraduate level.
    Reference | Related Articles | Metrics
    A multi-dimensional evidence dynamic trust computing model based on multi-agent
    JIANG Wei-jin, XU Yu-hui, GUO Hong, XU Yu-sheng
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2015, 50 (01): 1-11.   DOI: 10.6040/j.issn.1671-9352.2.2014.424
    Abstract589)      PDF(pc) (2506KB)(557)       Save
    The trust assessment is the core content of the trust management model based on the behavior. In order to complete the evidence sources of the trust assessment, a trust assessment model based on multi-dimensional evidence was proposed by using the trust management thinking. And this model introduces information on operating behaviors level of the main network into the traditional trust assessment model which only considerates feedback information of transaction, and computes trust based on multi-dimensional evidence sources on feedback information of transaction and network operating behaviors level, and extends the evidence source, and overcomes the defect when trust is assessed only according to a single type of evidence source. In addition, the application of the improved DS evidence to synthesize multi-dimensional evidences can solve the problem of uncertainty of evidences.
    Reference | Related Articles | Metrics
    Bifurcation analysis and control of HR neuron model under electromagnetic induction
    QIAO Shuai, AN Xin-lei, WANG Hong-mei, ZHANG Wei
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2020, 55 (9): 1-9.   DOI: 10.6040/j.issn.1671-9352.0.2019.634
    Abstract178)      PDF(pc) (2693KB)(248)       Save
    It is of great value to study the firing activity of neurons under electromagnetic radiation for the control and treatment of neuron-related lesions. Based on the theoretical analysis and numerical simulation, the bifurcation structure of HR neuron model of magnetic flux and its stability control of subcritical Hopf bifurcation are studied. Through the numerical simulation, it is found that the system has alternations of plus period 1 bifurcation, double period bifurcation and chaos in the two-parameter region. In addition, through the theoretical analysis of distribution and stability of the system equilibrium point under the change of external stimulus current, it is concluded that there exists a supercritical Hopf bifurcation point in the system, and a hidden limit cycle attractor is found near the subcritical Hopf bifurcation point. By using Washout controller, the subcritical Hopf bifurcation stability control is realized, which eliminates the phenomenon of hidden discharge and helps reveal and understand the internal mechanism of the generation and transformation of hidden discharge in neurons.
    Reference | Related Articles | Metrics
    Reversible data hiding in homomorphic encrypted domain based on Paillier
    Min-qing ZHANG,Neng ZHOU,Meng-meng LIU,Han WANG,Yan KE
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2020, 55 (3): 1-8,18.   DOI: 10.6040/j.issn.1671-9352.2.2019.021
    Abstract281)   HTML123)    PDF(pc) (5044KB)(247)       Save

    Based on the homomorphism and probability characteristics of Paillier cryptosystem, a reversible data hiding in homomorphic encrypted domain based on Paillier is proposed. First, the original image is divided into blocks. The pixel groups randomly selected by the image owner have one reference pixel and eight target pixels. Least significant bits (LSBs) of reference pixel and all bits of target pixel are self-embedded into other parts of the image by a method of RDH. In order to avoid overflow when embedding data, LSBs of the reference pixel are reset to zero before encryption. The encrypted reference pixel replaces the target pixels surrounding it, thereby constructing mirror central ciphertext. In a set of mirroring central ciphertext, the data hider embeds data in the LSBs of the target pixels by homomorphic addition, while the reference pixel remains unchanged. The receiver can directly extract the extra information by modular multiplication inverse between the target pixels and the corresponding reference pixel or directly extract the secret information by plaintext subtraction and can restore the original image without loss. The experimental results show that the average embedding rate of the algorithm is 0.18 bpp (bit per pixel) under the premise of ensuring the quality of the directly decrypted image.

    Table and Figures | Reference | Related Articles | Metrics
    Three-way decisions model based on the optimal center covering algorithm
    LIU Guo-tao, ZHANG Yan-ping, XU Chen-chu
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2017, 52 (3): 105-110.   DOI: 10.6040/j.issn.1671-9352.4.2016.216
    Abstract430)      PDF(pc) (1102KB)(583)       Save
    The three-way decisions model is the development of the traditional two-way decisions, and its three decisions include positive, negative, and boundary regions. The model widely used in the uncertain or incomplete information areas. The three-way decisions model is based on constructive covering algorithm(CCA)and it could automatic obtain three regions. However, the existing CCA is an uncontrollable random process with the covering center selected, that lead to the three-way decision classification accuracy uncertain. Thus we propose a novel three-way decision model to select the optimal center in constructive covering algorithm(OCCCA). The OCCCA model combines the nearest mean theory, obtains the mean of the one class in the data set, and then chooses the sample that nearest mean as the center. The experimental result shows that our methodology can improve about 5% than traditional CCA in the three-way decisions models classification accuracy.
    Reference | Related Articles | Metrics
    A kind of adaptive species optimization of wireless Mesh network QoS routing algorithm
    WAN Zhi-ping1, L Zhi-min1,2 *
    J4    2013, 48 (09): 10-16.  
    Abstract388)      PDF(pc) (1466KB)(1766)       Save

    For the network latency that is caused by the inequalities of wireless Mesh network bandwidth and load energy, and low operation speed of the routing algorithm and so on, a kind of adaptive species optimization of Wireless Mesh Network QoS routing algorithm (AQPSO) was proposed. To avoid the transmission delay and congestion of the data, the algorithm uses route evaluation function to search the best node path and through ant pheromone updating rules to balance the network load. Meanwhile, combined with the speciation policy of quantum behavior particle swarm optimization algorithm, a best position field adaptive search pattern was put forward, which improved the convergence speed of the algorithm. Viewing from the network latency and the algorithm convergence speed, the simulation results show that the proposed algorithm has a better effect than the improved ant colony QoS routing algorithm and QoS routing algorithm based on genetic algorithm.

    Related Articles | Metrics
    Improved certificateless aggregate signature scheme
    HOU Hong-xia1,2, ZHANG Xue-feng2, DONG Xiao-li2
    J4    2013, 48 (09): 29-34.  
    Abstract581)      PDF(pc) (693KB)(1704)       Save

    Recently, the aggregate signature is a hot topic concerned. An improved cerificateless aggregate signature scheme is constructed based on the certificateless cryptography. Compared with the existing schemes, the scheme has constant pairing computations, and the verification algorithm only requires three pairing computations, which greatly improves the efficiency of the signature verification. The improved scheme can resist the attack of TypeⅠand TypeⅡadversary in certificateless adversary model,and it is existentially unforgeable under the computational DiffieHellman assumption.The security proof of the improved scheme is given in the random oracle model.

    Related Articles | Metrics
    A new method of solving the assignment problem based on the permissible-edge algorithm of minimum cost flow problem
    XIONG De-guo, HU Yong-wen
    J4    2012, 47 (3): 103-109.  
    Abstract675)      PDF(pc) (2190KB)(1709)       Save

     A new algorithm of the assignment problem is proposed by constructing its minimum cost maximum flow model and applying the permissible-edge algorithm based on the principle of duality to the model. The new algorithm gradually expands the permissible network in the capacity-cost network by means of modifying the potential of labeled nodes subject to complementary slackness condition, and then augments flows on the permissible network, which proceeds until the minimum cost maximum flow of the original capacity-cost network is obtained. The non-zero edge of this maximum flow corresponds to the optimal solution of the assignment problem. During the iterating process, successive iteration will fully use the information of previous ones, which effectively reduces the computation. For non-standard assignment problems, this algorithm can be directly applied without converting the problem to the standard form.

    Related Articles | Metrics
    Computational humor researches and applications
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2016, 51 (7): 1-10.   DOI: 10.6040/j.issn.1671-9352.0.2016.266
    Abstract1523)      PDF(pc) (1413KB)(886)       Save
    Humor, as a special phenomenon of human communications, can warm up the atmosphere and eliminate embarrassment. In recent years, with the research development of artificial intelligence, research area related to how to model humorous expression using computers becomes a hot topic in natural language processing tasks, and evolves to become a new subject, called computational humor. Computational humor aims to recognize and interpret humorous expressions in context using natural language processing technologies, and construct humor based computational models. In this article, we firstly introduce the backgrounds of computational humor research and detail the reasons for which humor can be modeled using computers. After that, we review related research in two lines, humor recognition and humor generation, and give the computational procedure of them respectively. Finally, we introduce some applications of humor computing in different tasks, including chatting robots, machine translation, children teaching software and English teaching. Overall, we review the recent research work in the area of humor computing to motivate new ideas and broaden horizons for further research in this area, which can help computers understand the natural language of humans, and promote the development of artificial intelligence.
    Reference | Related Articles | Metrics
    A LBS privacy-preserving method for geo-fencing services
    YANG Song-tao, MA Chun-guang, ZHOU Chang-li, ZHANG Zong-li
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2014, 49 (09): 69-73.   DOI: 10.6040/j.issn.1671-9352.2.2014.065
    Abstract696)      PDF(pc) (1753KB)(1117)       Save
    Geo-fencing technology is widely used in recommender systems, advertising push and other services. It has serious risk of personal privacy leakage threat. Focus on triggered query in Geo-fencing, the theories of the secure tripartite computation were applied in the research of the LBS privacy-preserving model, which follows the concept of location obfuscation and location k-anonymity and reaches the target of user's identity unlink-ability and location's untraced-ability. Location anonymity hidden problem and secret authentication problem were solved by computational geometry techniques and principles of cryptography in the triggered queries scenario. Theoretical analysis shows the model is security. From the perspective of privacy-preserving, the attackers do not get new knowledge from LBS system. Experiments demonstrate the computational efficiency and communication loads of this medel are superior to the traditional spatial-temporal cloaking methods.
    Reference | Related Articles | Metrics
    Mean model based IBCF algorithm
    QI Li-li, SUN Jing-yu*, CHEN Jun-jie
    J4    2013, 48 (11): 105-110.  
    Abstract6208)      PDF(pc) (1451KB)(581)       Save

    The item-based collaborative filtering algorithm (IBCF),a recommendation algorithm with high precision,simple and easy to use in actual system, is widely used in the field of recommendation systems. But it meets a higher computational time complexity for similar calculation because of the long length of item vector. In this paper, a sampled approach firstly is suggested to represent an item vector called mean model item vector representation through analyzing theory of IBCF algorithm, to solve the problem of the long length of item vector and cut down the computational time. Experiments using Movie Lens datasets show that the algorithm is very efficient to cut down the computational time on the premise of accuracy. Furthermore, some right sampling methods can be used to optimize the calculation method of similarity in order to meet practical application requirement.

    Related Articles | Metrics
    Privacy preservation for location-based services
    Hai-yan KANG,Wan-xiang ZHU
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2018, 53 (11): 35-50.   DOI: 10.6040/j.issn.1671-9352.0.2018.128
    Abstract955)   HTML27)    PDF(pc) (1792KB)(559)       Save

    The disclosure of privacy involves not only the users' locations and queries data, but also the users' identities, interests, social habits, behavior patterns and privacy hidden in these data. A review of privacy preserving for location-based services (LBS) is provided. The application scenarios of location services is introduced, the method of location privacy attack is given. The existing research on the privacy protection system structure and protection technology is summarized. Finally, future research is prospected, and a cache-based spatiotemporal disturbance method and LBS privacy protection metric hypothesis scheme are proposed.

    Table and Figures | Reference | Related Articles | Metrics
    Application of a new accelerating Bregman iterative algorithm in the sparse least squares problems
    WANG Wen-shu, LI Wei-guo, QIN Shu-lan
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE)    2016, 51 (6): 92-98.   DOI: 10.6040/j.issn.1671-9352.4.2015.004
    Abstract414)      PDF(pc) (907KB)(696)       Save
    Based on the residual Bregman iterative algorithm and fixed point iteration, a novel Bregman iterative algorithm is proposed. Then, inspired by the equivalence of the linearized Bregman and the gradient descent algorithm of the dual problem, a new algorithm by introducing Nesterov accelerating technique into the predictor-corrector method is put forward for solving the sparse least squares problems. Simultaneously, It is proved that the solution sequence obtained by the new method is the optimal solution of the sparse least squares problems. Finally, we use the new method to the sparse signal recovery problem. The numerical results show that the new method is faster and more efficientthan the old ones.
    Reference | Related Articles | Metrics