Computer and Modernization

Homology Analysis of Ransomware Based on Sequence Alignment

GONG Qi, CAO Jin-xuan, LU Tian-liang

2018, 0(02): 1-5. doi:10.3969/j.issn.1006-2475.2018.02.001

Asbtract ( 160 )

References | Related Articles | Metrics

The number of ransomware is increasing rapidly while few belong to new family, most of them are mutations. A new homologous analysis approach based on API sequence of ransomware is proposed. The paper uses sandbox to extract ransomwares dynamic behavior for analyzing API category, and then encodes the feature as well as removes the repetition. Also, the sequence alignment algorithm is used to calculate the similarity between different ransomware. The dataset for the experiment contains 6 different families of ransomware and their variants. The result shows that proposed method performs well in analyzing the homology of ransomware which can be used to distinguish unknown software.

Blind Recognition for STBC with Transmission Impairments Based on Cyclostationarity Test

FANG Wei, YAN Wen-jun, LING Qing, ZHANG Li-min

2018, 0(02): 6-11. doi:10.3969/j.issn.1006-2475.2018.02.002

Asbtract ( 134 )

References | Related Articles | Metrics

An algorithm is proposed based on fourth-order cyclostationarity to classify Spatial Multiplexing (SM) and Alamouti space-time block code (STBC) with transmission impairments when a single antenna is employed at the receiver. Firstly, we prove the feature that Alamouti STBC signals hare cyclic frequency. Secondly, the fourth-order cyclic cumulants of the received samples are calculated to observe the cyclic frequency. Thirdly, we exploit the discriminating features provided by cyclic frequency of different STBCs, and construct cyclostationarity test to detect the cyclic frequency to achieve the classification. The algorithm can work in single received antenna scenarios. Monte Carlo simulation shows that the algorithm does not require the channel state information, modulation of the transmitted signal, carrier phase and time offset. It does not need to accurately know the carrier frequency offset and performs well.

A Mass Diffusion Recommender Algorithm with User Trust Network

LIANG Yi, HAN Li-xin

2018, 0(02): 12-16. doi:10.3969/j.issn.1006-2475.2018.02.003

Asbtract ( 162 )

References | Related Articles | Metrics

In the recommend systems, modeling resource is a vital issue. Based on the study of the traditional user-based resource models, we find that it is always assumed that users are independent of each other, trust relations among the users are not fully used, leading to strong Matthew effect, cold start problem and so on. This paper designs a common user trust network based on the tag co-occurrence, upon which the PageRank algorithm is used to refine the resource model. Further, an improved diffusion process is performed to get the recommend results. Compared with the previous algorithms, experimental results show that our algorithm significantly improves accuracy， recall and F1-measure of recommendations.

An Improved FP-Growth Algorithm in Planning of Leisure Travel

ZI Yun-fei, LI Ye-li, SUN Hua-yan, ZHANG Li-jing

2018, 0(02): 17. doi:10.3969/j.issn.1006-2475.2018.02.004

Asbtract ( 152 )

References | Related Articles | Metrics

According to the disadvantage of consuming great memory cost in storing massive data existing in the FP-Growth algorithm of mining association rules, the paper presents an improved algorithm which adds interestingness to FP-Growth, and then compares it with Apriori and FP-Growth algorithm, the improved algorithm greatly reduces the memory cost and improves the efficiency of system execution. Based on the idea of combining the improved algorithm with the tourism route planning, taking Yunnan tourism as the tourism planning object and fully applying the large data of the tourism website, the paper designs a mining system for tourism route planning, and finds out the association rules between tourist routes and scenic spots.

Time Series Similarity Search Based on Relevance Feedback

LIU Qi, ZHANG Peng-cheng, WANG Ji-min

2018, 0(02): 22-26.

Asbtract ( 184 )

References | Related Articles | Metrics

The traditional time series similarity search based on relevance feedback is to combine positive feedback with negative feedback to create new query vectors. This does not make full use of the value of negative feedback sequence, and it is easy to make too many changes to the initial query vector. This paper proposes a similar time series search method based on relevance feedback. The positive relevance feedback and negative relevance feedback are carried out separately. This way makes the results far away from negative relevant sequence. The results of similarity search on UCR data sets show that the similarity search method based on relevance feedback can improve the accuracy of similarity query.

POS Maintenance Information Service Platform Based on GIS

HOU Hai-xu， ZHOU Xiao-hui

2018, 0(02): 27. doi:10.3969/j.issn.1006-2475.2018.02.006

Asbtract ( 132 )

References | Related Articles | Metrics

Aiming at the problem of efficient maintenance service of POS machine equipment in commercial banks, this paper develops the POS maintenance information service platform. The platform uses B/S architecture, the background uses Spring MVC and MyBatis integration framework, front-end uses Bootstrap and easyUI frameworks and geographic information system(GIS) path optimization algorithm for the design and implementation. The each repair information of customer side can be processed and be sent to maintenance timely. It makes the whole repair and maintenance information service process automatic, intelligent and standardized, and improves efficiency effectively.

Design of Child Anti-lost System Based on NB-IoT

ZHANG Zhong, ZHU Tian-tian, MA Xing-lu

2018, 0(02): 30. doi:10.3969/j.issn.1006-2475.2018.02.007

Asbtract ( 222 )

References | Related Articles | Metrics

There are many children lost in China every year, once child lost, the chance of getting back is very little. Based on this, the design of the childrens anti-lost system based on narrow band Internet of Things (NB-IoT), adopts NB-IoT insteading of traditional Bluetooth, WiFi and GPRS, has the advantages of low power consumption, wide coverage, strong penetration, good signal etc. Using Beidou positioning module to replace the traditional GPS has the advantages of no blind spot and accurate positioning. Using NB-IoT to establish a connection with the base station, the positioning information output by Beidou satellite navigation system is transmitted to the base station through NB-IoT. Users can view the location information of children at any time through the mobile client to help parents quickly locate and retrieve children and reduce the chance of child loss.

Application of WeChat Public Service Platform in E-government

CHU Jia-qi1,2, LIU Cong-jun1,2

2018, 0(02): 33. doi:10.3969/j.issn.1006-2475.2018.02.008

Asbtract ( 276 )

References | Related Articles | Metrics

With the popularity of smart phones, WeChat and other social APPs gradually become the main tools for the public to access to news, information and services instead of traditional website. This paper proposes and implements a scheme of e-government service platform based on the WeChat public account. Furthermore, the overall performance of the system is optimized by using browser local cache and CDN to improve the speed of response to user access and page loading. The system provides efficient government information query and portable online functionality of government services, thus to ensure timely disclosure of government information, to improve government information construction standard and public service capability.

Incremental Updating Algorithm of Application System

HUANG Lin, YANG Jun， XU Liang-liang

2018, 0(02): 39. doi:10.3969/j.issn.1006-2475.2018.02.009

Asbtract ( 168 )

References | Related Articles | Metrics

Aiming at the efficiency issues which the application system upgrade brings, such as high service pressure caused by large-scale updating and large amount of network data transmission, we propose an application system construction model based on the micro-service architecture. Then, to update the application system locally, based on BsDiff and other traditional binary file incremental updating algorithm, the incremental update algorithm of application system-level ASIUpdate is proposed. Finally, the prototype software of the new algorithm is developed. Comparing the update efficiency with the traditional application system version, the test results show that the proposed algorithm has obvious improvement in efficiency and resource consumption.

SEIR Microblog Public Opinion Communication Model with Positive and Negative Feedbacks

QIU Xiu-lian1， TIAN Xiao-hu1,2， LIAO Wen-jian1

2018, 0(02): 44. doi:10.3969/j.issn.1006-2475.2018.02.010

Asbtract ( 183 )

References | Related Articles | Metrics

The social network public opinion has become the main position of social public opinion. Because the traditional public opinion prediction model is difficult to describe the real spreading process of social network, a social network topic propagation model based on infectious disease dynamics with positive and negative feedbacks is proposed. This paper analyzes the real characteristics of social network public opinion topic, adds two significant features: Internet mercenaries and zombie fans as positive and negative feedback in the dissemination of public opinion topics, which promote and inhibit the spread of opinion topics respectively. The SEIR microblog pubic opinion communication model with positive and negative feedbacks is constructed, which improves the public opinion prediction accuracy compared with traditional models, and gets the influence of positive and negative feedback on public opinion communication.

A Popularity Prediction Algorithm Based on Video Characteristics and Historical Data

ZHAO Ming-yan， LI Ze-ping

2018, 0(02): 49. doi:10.3969/j.issn.1006-2475.2018.02.011

Asbtract ( 163 )

References | Related Articles | Metrics

For the popularity prediction of streaming media, a model of popularity prediction based on video characteristics and historical data is proposed. Firstly, according to the video characteristics and the influence in the social network, the popularity of video is predicted using K-Nearest Neighbor(KNN). Subsequently, based on the results of last step, combined with the Autoregressive Moving Average (ARMA) model, the on-demand quantity of the video is predicted using historical data. Finally, the experiment is carried out by crawling the Douban film and Sina microblogging data. The results show that the recall rate of the model is higher than that of the Naive Bayesian classifier, and the average square root error ( RMSE) is decreased by about 20%, compared with the ARMA model.

Research on Parameters Setting and Classification Characters of Four Classification Algorithms

WANG Zheng-jie, YANG Wei-li, WANG Zhe, HOU Yu-shan, GUO Yin-jing

2018, 0(02): 54. doi:10.3969/j.issn.1006-2475.2018.02.012

Asbtract ( 174 )

References | Related Articles | Metrics

The quantitative research and analysis about the classification algorithm are not often sufficient to choose the suitable algorithm. In this paper, the K-Nearest Neighbor algorithm, SVM and decision tree are quantitatively analyzed about classification accuracy and running time by using different the parameters of the algorithm, the data noise and the number of nodes. Firstly, parameters effects of these algorithms are studied. Then, the optimal parameters are selected to analyze the influence of different noise on the classification accuracy. At last, the influence of the number of nodes on the classification accuracy and the running time is analyzed. The Scikit-learn module is used to simulate the content of the discussion. The experimental results clearly show the classification characteristics of these classification algorithms under different parameters, which provide guidance for the classification of the actual data.

Tree Modeling Method Based on Principle of Plant Apical Dominance

WANG Chun-hua, HAN Dong

2018, 0(02): 61. doi:10.3969/j.issn.1006-2475.2018.02.013

Asbtract ( 158 )

References | Related Articles | Metrics

Based on plant’s self-organizing modeling thought, a novel modeling method of trees’ basic growth unit—bud competing for light is proposed. This method consists of three parts: growth environment computing, resource allocation and growth simulation, which is easy to understand and operate. The experimental simulations show that the generated image is lifelike and natural, meanwhile apical dominance, geotropism and phototropism phenomena can be simulated just by several parameters. This paper provides a new idea for trees’ modeling in virtual world.

Research on Improvement of Feature Weights in Text Classification

LI Peng-peng， FAN Hui-min

2018, 0(02): 66. doi:10.3969/j.issn.1006-2475.2018.02.014

Asbtract ( 132 )

References | Related Articles | Metrics

In order to overcome the shortcomings of traditional TF-IDF (Term Frequency Inverse Document Frequency) algorithm, the improved TF-IDF-dist algorithm is proposed by using the distribution of feature words. The experimental results show that the improved algorithm has an average increase of F1 value by 3.2% in the different feature dimensions. With the different feature selection algorithm, the F1 value is increased by 2.75% and the improved TF-IDF-dist algorithm has more adaptability on the imbalance datasets. It shows the validity of the algorithm in text classification.

A Flexible Parallelization Method for Elliptic Curve Cryptosystems

WU Ke-ke, HUANG Guo-wei, KONG Ling-jing

2018, 0(02): 71. doi:10.3969/j.issn.1006-2475.2018.02.015

Asbtract ( 142 )

References | Related Articles | Metrics

This paper proposes a flexible parallel method of scalar multiplication for elliptic curve cryptosystems (ECC) based on the proposed scalar partition and integration models. Focusing on parallelizing ECC scalar multiplication operations at the scalar multiplication algorithm level, the proposed method can be implemented into various parallel systems. In contrast to previous parallel scalar multiplication methods, the proposed method is flexible. Furthermore, the time complexity of the proposed parallel scalar multiplication method can be reduced to (logk)A+kD. The optimal time complexity of the proposed method is reduced about 30% compared with classic binary method by an example.


Improved Collaborative Filtering Recommendation Algorithm Based on ALS Model

NI Man-man1,2

2018, 0(02): 76. doi:10.3969/j.issn.1006-2475.2018.02.016

Asbtract ( 158 )

References | Related Articles | Metrics

The recommendation system can provide personalized recommendation services to users based on the user’s basic information and behavior analysis. Therefore, the recommendation system has become a research hotspot in recent years. This paper studies on the algorithm of collaborative filtering recommendation based on ALS model. The implementation of the algorithm uses a distributed platform, and the experimental results show that compared with the previous single-node implementation, the proposed algorithm has greatly improved the computational speed. In addition, this paper reduces the attribute information loss of invisible factor on the loss function, and introduces the interest forgetting function in the predictive score obtained by the optimal model. The experimental comparison shows that the optimized algorithm effectively improves the accuracy of the recommended system.

A Solution of Multiple Target and Random E-examination Paper Based on Random and Reduce Range Algorithm to Strategy Set

ZHOU Hua-jun, DING Ai-fen, LYU Xiao-jun

2018, 0(02): 80. doi:10.3969/j.issn.1006-2475.2018.02.017

Asbtract ( 106 )

References | Related Articles | Metrics

Research on generating test paper in electronic examination system (E-Exam System) remains a comprehensive problem in teaching and examination reform of colleges and universities because it requires high efficiency, randomness and balance in algorithm, meanwhile the extracting test paper combination should meet the users’ custom sets strategy. This paper presents a strategy-sets-based reduce range of algorithm’s parameters, which firstly applies a random shuffle algorithm towards test papers and the process of generating test papers to guarantee the randomness of abstracting test questions; then prunes the value domain ahead of time by using strategy sets and reduces the value range of the current target set to ensure fast return of algorithm; finally rotates the abstracting targets with strategy sets to realize test paper adjustment based on strategies. After the test of abstracting 1000 test papers from a 100000 questions Test Bank, this algorithm shows the following features: it can abstract item quantity accord to the goal within very short time (127 s); algorithm efficiency and the number of test paper composition show a variation trend of quadratic function; it has less sensitiveness towards the number of questions in test bank so this algorithm is suitable for a relatively large test bank.

A Multi-thread Asynchronous Dispatch Algorithm for Large-scale Virtual Terrain Data

REN Zi-jian1, CHEN Lu2

2018, 0(02): 84. doi:10.3969/j.issn.1006-2475.2018.02.018

Asbtract ( 109 )

References | Related Articles | Metrics

A real-time dispatching algorithm for massive terrain data is proposed, in order to implement large-scale terrain rendering. On the basis of traditional quadtree model, the algorithm is used to make further organization and index for the data. Based on this, high performance spatial querying is archived by the use of Hilbert storage indexing algorithms. And then, an asynchronous dispatch mechanism is designed based on I/O Completion Port (IOCP) and multi-threaded technology for optimally I/O operations, data loading and unloading. Finally, we carry out a comparative analysis by the use of the algorithm and single thread synchronous I/O algorithm. The results of the experiments show that, compared to traditional algorithms, the proposed algorithm has efficient data scheduling performance and can meet the needs of real-time rendering of massive terrain environment.

Sentiment Analysis of Chinese Microblog Based on Expand-dictionary and Semantic Rule

LI Ji-dong， WANG Yi-zhi

2018, 0(02): 89. doi:10.3969/j.issn.1006-2475.2018.02.019

Asbtract ( 145 )

References | Related Articles | Metrics

Firstly this paper focuses on the occurrence rule of new words in microblog texts, finds microblogging new words through adverbs, then calculates the SO-PMI between the new words and the emotional benchmark words by optimized PMI algorithm, based on which the new words are divided into praiseful and derogatory categories and then been added to microblog domain dictionary. Secondly, basic emotional dictionary is constructed, considering the uniqueness of microblogging text and the characteristics of Chinese language, we construct microblogging expression dictionary, negative word dictionary, adverbs dictionary, conjunctions dictionary. Finally, combined with the emotional dictionary and semantic rules, we carry on an emotional analysis on Chinese microblogging by the means of emotional weighting with microblogging expressions. The validity of the proposed analysis strategy is verified by testing the microblogging data set.

An Attention-based C-GRU Neural Network for Text Classification

YANG Dong, WANG Yi-zhi

2018, 0(02): 96. doi:10.3969/j.issn.1006-2475.2018.02.020

Asbtract ( 212 )

References | Related Articles | Metrics

Text classification is the classical research direction in NLP and plays an important role in information processing. At present, deep learning network has achieved the remarkable performance in image recognition, machine translation and other fields and it also has been proved to be capable of learning higher-level sentences and document representation in NLP tasks. In this paper, based on GRU model and the convolutional layer in CNN, we propose a novel hybrid text classification model called Attention-based C-GRU. Moreover, we introduce Attention model in our model, which effectively highlights the role of key words and optimizing the extraction of features. We leverage the model to learn the meaning of text and evaluate it on topic classification, question classification and sentiment classification tasks. The experiment demonstrates the effectiveness of our approach in comparison with baseline models and state-of-art methods.

Role Identification of Money Laundering Based on Improved Random Forest

ZHANG Hao, HUANG Wei, HU Guo-chao

2018, 0(02): 101. doi:10.3969/j.issn.1006-2475.2018.02.021

Asbtract ( 128 )

References | Related Articles | Metrics

Aiming at the problems of low accuracy and delay of the method of identifying the bank account in the money laundering transaction, the method of public security unit handling, and the existing research based on machine warning, this paper analyzes the trading behavior characteristics of the bank and the clients in the money-laundering crime group, extracting a series of features like the personal background attributes, trade statistics, trading platform, trading behavior to depict outliers from four kinds of view. The features are selected and optimized by random forest model. The annotated data is trained and verified. An application is formed for automatic recognition of traders involved in money laundering. Through the actual data validation, those banking operators with serious hazards can be found out.

A Face Verification System on Mobile Terminal Based on Depth Learning

LIU Cheng1, TAN Xiao-yang1,2

2018, 0(02): 107. doi:10.3969/j.issn.1006-2475.2018.02.022

Asbtract ( 136 )

References | Related Articles | Metrics

As a new authentication technology, face verification is widely used in access control, attendance and other needs of the occasion of authentication. This paper designs and implements a face verification system based on the Android system platform, taking into account the requirements and production environment of mobile face verification and the efficiency and portability of existing face verification algorithms. The system can be deployed off-line in the mobile device equipped with Android system, through the camera obtaining a face image and the local image processing data to complete face verification work. In the algorithm, the system uses deep convolution neural network for image processing and face feature vector extraction to improve the accuracy of face verification. In the implementation, through the joint compiler Java and C+〖KG-*3〗+ codes improving the efficiency of the algorithm to adapt to the depth learning algorithms in the mobile side of the application. Experiments show that the system can quickly ensure the accuracy rate of 97.16% under the premise of rapid completion of the face verification process, basically meet the needs of industrial applications.

A Deep Learning Face Recognition Algorithm Based on Local Ternary Pattern

ZHENG Qiu-mei， XIE Huan-li， WANG Feng-hua， SU Zheng， LIU Zhen

2018, 0(02): 112. doi:10.3969/j.issn.1006-2475.2018.02.023

Asbtract ( 144 )

References | Related Articles | Metrics

In order to solve the problems of the heavy dependence on artificial selection during the process of traditional feature extraction and leaving local features out of consideration in traditional DBN network， this paper proposes a face recognition algorithm based on local ternary pattern and deep learning (LTDBN) to get higher face recognition ratio. This algorithm firstly segments a normalized face image into multiple small parts equally and carries out LTP algorithm for each part. Then the histogram is used to get the final image features. These image features are served as the input data of DBN. The greedy learning algorithm trains and recognizes the whole network level by level. The recognition ratio reaches 98.75%， 100% and 96.62% respectively in public face databases including ORL， Yale and Yale-B. The experiment results indicate that LTDBN algorithm is markedly superior to other existing algorithms in recognition ratio and mitigates the negative effects of factors such as illumination and posture.

3D Measurement Based on Combination of Structured Light and Binocular Vision

ZHANG Yong-ju1, GU Xu-bo1, ZHANG Jian1, WANG Bing2, GUO Ling2

2018, 0(02): 118. doi:10.3969/j.issn.1006-2475.2018.02.024

Asbtract ( 192 )

References | Related Articles | Metrics

Among three-dimensional measurement technologies, traditional binocular stereo vision requires a large amount of computation and has low matching accuracy for images without distinctive features. The combination of 3D measurement technology based on structured light and binocular imaging technology is adopted to solve the difficulty in feature point searching and improve the accuracy of measurement. Compared with the traditional structured light imaging technology, the proposed method only needs a conventional dual stereo camera calibration instead of the projector calibration, which reduces the complexity of the system and improves its maneuverability and flexibility. Finally, the 3D measurement experiment is carried out and the experimental results show the feasibility of the technology.

Image Semantic Segmentation Based on Region Proposal Network

YANG Zhi-yao1,2, PENG Zhao-yi1,2, WEN Zhi-qiang1,2

2018, 0(02): 122. doi:10.3969/j.issn.1006-2475.2018.02.025

Asbtract ( 139 )

References | Related Articles | Metrics

In order to improve the problems of roughness and lack of detail in the semantic segmentation of image, a joint network structure is proposed, which combines the regional proposal network and realizes the convolution layer sharing. The regional proposal network is used to generate some regional proposal boxes which contain category information. The regional proposal boxes are used to correct the segmentation results of the fully convolutional semantic segmentation network. Experiments show that this method can effectively improve the classification accuracy of pixels and get better segmentation results.

Table of Content