Table of Content

28 February 2015, Volume 0 Issue 2

Previous Issue    Next Issue

Research Method of Web Collective Intelligence in Era of Big Data#br# #br#

WANG Hua1,2, ZHAO Dongjie2,3, YANG Haitao3, ZHAO Hongli3, LI Zhi3, LI Shenlong1

2015, 0(2):  1.  doi:10.3969/j.issn.10062475.2015.02.001

Asbtract ( 299 )  

References | Related Articles | Metrics

Firstly, the internet environment in the era of big data where the crowds interact adequately each other was analyzed. Secondly, the concept of web collective intelligence was put forward and explained, and its characteristics involving network data driven, complex interaction form, strong network effect, knowledge production dominated and cognition with uncertainty were obtained, then the research paradigm of web collective intelligence was put forward, the method with complexity science methodology as the guidance adheres to syncretism, took complexity science, networked data mining and artificial intelligence with uncertainty as backbone theory and method, outstood web collective intelligence characteristic and interdisciplinary research, and applied a technique with a combination of system analysis, modeling analysis and simulation analysis to study the scientific problems of web collective intelligence from multiscale and multilevel perspective, which solves the deficiency problem of the research theories and methods of web collective intelligence and deepens the knowledge of web collective intelligence and social computing.



Recombination and Adaptive Technology of Responsive Web Pages Based on HTML5

JIANG Lingyan， ZHA Yinghua

2015, 0(2):  7. 

Asbtract ( 193 )  

References | Related Articles | Metrics

In webpage browsing and Web accessing, through the comparison between mobile terminal and PC, it’s found that the display screen size is not uniform, the layout of the page does not adapt, and the noise information is too great. So a great quantity of website on network is need to transform and recombine the webpage information adaptively when accessed by the mobile terminal. The new label and features provided by HTML5 can better adapt to the multiterminal access needs, the type of response page design principles in the mobile terminal access can also provide users with better service and experience. This paper presents a way which can translate existing website and webpage into HTML5based response page by using framework of foreground technology to recombine the webpage information adaptively. The website can be better adapted to the environment of multiterminal access, provide better service for user.



Androidbased Google Map Location Software

FENG Yuping1, WANG Shuguang2

2015, 0(2):  11. 

Asbtract ( 167 )  

References | Related Articles | Metrics

For locationbased services are widely used in tourism, query line, mobile location services and other occasions, using Android rich APIs, by applying Google Map API Key to call Google Map realtime, this paper designed and implemented a mobile terminal positioning software which can query latitude and longitude, while supporting switching map zoom, view features. Test results show that the software can accurately locate the target position, flexibly switch views and zoom view, and provide users with accurate and convenient location services.


Combined Sales Prediction Model Based on ARMA and BP_AdaBoost

YAN Bo1, ZHOU Zaijin1, LI Guohe2,3, QI Jia1

2015, 0(2):  14. 

Asbtract ( 179 )  

References | Related Articles | Metrics

In order to improve the accuracy of prediction, a combined sales prediction model is established. The historical sales data is nonlinear, timevarying time series. It consists of two parts, the linear and nonlinear. By using the ARMA model, the linear part can be predicted while the nonlinear part can be predicted by using BP_AdaBoost model. Then the two prediction results are added together. The combination model overcomes the problem of low accuracy by using ARMA model alone. What’s more, it also overcomes the problems that BP neural network model is easy to fall into local minimum. The experiments show that the combination model can improve the accuracy of sales prediction and reflect market rules more accurately and comprehensively.



Cost Sensitive Relevance Vector Machine

SU Lequn， FENG Aimin

2015, 0(2):  19. 

Asbtract ( 199 )  

References | Related Articles | Metrics

Relevance Vector Machine (RVM) is a sparse model proposed on the basis of sparse Bayesian framework, it has been widely studied and applied in the field of machine

learning in recent years because of its strong sparsity and generalization ability. However, like the traditional decision tree, neural network algorithm and support vector

machine, RVM does not have the expense of sensitivity, can not be directly used for costsensitive learning. To deal with the cost sensitive problem brought by

misclassification in supervised learning, costsensitive relevance vector classification(CSRVC) algorithm was proposed by integrating misclassification cost of each type

sample based on RVM. Experiments show that CSRVC has good sparsity and can effectively solve the problem of costsensitive classification.


An Algorithm for Travel Route Planning Merging with Weather Monitoring and Nowcasting

WANG Xing, ZHU Weicong, ZHANG Hao, ZHANG Xiaoxiao

2015, 0(2):  25. 

Asbtract ( 130 )  

References | Related Articles | Metrics

On the basis of traditional GPS route planning and navigation, we present an algorithm for travel route planning merging with weather phenomena monitoring and nowcasting which affect driving safety such as the sleet fog, wind, rain and icy roads, etc.， and then illustrate the design of the logical structure of the system, the key business processes and physical deployment， finally check the effects of the algorithm through the integrated system which has been completed under Android platform. The algorithm and system not only are applied to the weather forecast and warning for shortdistance travel inside the city, but also suitable for intercity or interprovincial longdistance highway route planning, so as to effectively reduce the traffic accidents attributable to bad weather.

An Entropy Model to Infer Social Strength Based on Texts of Subjects

DENG Zhongsheng

2015, 0(2):  30. 

Asbtract ( 141 )  

References | Related Articles | Metrics

In social networks services, text information based on subjects can reflect users’ certain interest on the subjects. By analyzing the common interests between users, users’ social strength can be inferred, and it can be used to friend recommending and advertisement pushing. This paper introduces two independent ways: diversity and weighted frequency, to infer social strength between users. The parameters of the models are learned from the Katz score of dataset. The results of experiment show that this model can accurately infer the social strength between users.

A Solution of Largescale TSP Based on Improved Genetic Algorithm

LEI Yumei

2015, 0(2):  34. 

Asbtract ( 175 )  

References | Related Articles | Metrics

TSP not only describes the problem of travelling around a number of cities, but also stands for a number of problems in some other fields. Thus, it is meaningful to find an effective solution to largescale TSP. As for the solution to find the minimum distance of a loop consisting of a large number of cities in TSP， in this paper, we propose such a new solution based on genetic algorithms. It adopts the idea of dividing and ruling, and utilizes a genetic algorithm based on improved initialization method and genetic operators to improve its performance. Experimental results across multiple datasets illustrate the proposed algorithm performs well in optimizing convergence result and solving the problem of premature convergence to some extent.

An Evaluation Method for ERP Selection Based on Measure of Medium Truth Degree

ZHANG Gui, ZHANG Yuping, CHEN Haiyan

2015, 0(2):  40. 

Asbtract ( 159 )  

References | Related Articles | Metrics

ERP selection is the key problem of the enterprises’ successful implementation of the ERP software. A comprehensive evaluation method for ERP software based on measure of medium truth degree is proposed. And this method emphasizes the horizontal contrast between the software to be evaluated, and then the underlying index score will be mapped to the axis so as to calculate the quality of the index with the distance ratio function, as well as implement ERP evaluation from qualitative description to quantitative calculation. Besides, the contrast experiment shows that this method is more simple and accurate.

Realtime Scheduling Algorithm Based on Multifeatures Synthesis

MAN Li1, ZHU Ruilong 2

2015, 0(2):  44. 

Asbtract ( 144 )  

References | Related Articles | Metrics

Currently, most of the realtime scheduling algorithms define the priority of task based on a single characteristic parameter. This paper presents a realtime scheduling algorithm based on the synthesis of multifeatures. It processes the specific highpriority task and does not have any impact on the scheduling of other tasks. When system overload it can effectively avoid the sharp decline of the EDF (Earliest Deadline First) algorithm performance. Experimental results show that the algorithm effectively guarantee the priority of specific task scheduling, and the performance has significantly been improved compared with EDF algorithm.

Semidynamic Integration Selection Classification Method

LI Rui1, YUAN Xiaoling2

2015, 0(2):  48. 

Asbtract ( 142 )  

References | Related Articles | Metrics

Traditional Dynamic Ensemble Selection (DES) in ensemble learning needs to select individual classifiers for all the test samples. However, it leads to highly computational cost. Due to this issue, a new Semi Dynamic Ensemble Selection (SemiDES) strategy is proposed in this paper, which consists of two stages. Individual classifiers are selected for all the test samples in the first stage. In the second stage, the classifiers for each test sample are selected dynamically. The final result is obtained by integrating the output of the two stages. The experimental results on UCI data set demonstrate the proposed method can obtain a better classification performance. Moreover, SemiDES can reduce the computational cost greatly.

Study on Pedestrian Detection Based on Improved HOG Eigenvalues

ZHANG Guining

2015, 0(2):  52. 

Asbtract ( 142 )  

References | Related Articles | Metrics

The pedestrian detection mechanism was proposed for solving these defects such as weak expression ability of HOG reduced by automatically ignoring differences in the opposite direction of its gradient direction when taking the current gradient direction histogram as descriptor for applying to pedestrian detection. Basing on analysis of the traditional HOG descriptor, the improved HOG descriptor was designed by connecting histograms; and a new normalization technique was proposed to embed into the improved HOG descriptor for enhancing its expression ability. The experiments results basing on several datasets show that: comparison with traditional HOG feature method, this proposed mechanism had higher accuracy and lower miss rate.

A Balanced Update Ant Colony Optimization for Aircraft Arrival Sequencing and Scheduling

LI Yuanzhen， YANG Qun， DUAN Xi

2015, 0(2):  57. 

Asbtract ( 154 )  

References | Related Articles | Metrics

Aircraft arrival sequencing and scheduling (ASS) is a key problem of air traffic control (ATC). According to the ASS model, a balanced update ant colony algorithm (BUACO) is proposed in this paper. BUACO balanced update the pheromone and enhance the global search ability of the algorithm by taking advantage of the difference between the current solution and the global optimal solution, in order to generate a better solution. The experiments show that BUACO’s performance can be increased by 12.9% with a shorter computation time than the comparison algorithms when solving ASS problem, which is conductive to arrange a suitable flight landing sequence based on realtime situation for ATC. 

Dynamic Replicas Strategy Based on Predicted Popularity

ZHANG Song1， DU Qingwei1， SUN Jing2， SUN Zhen2

2015, 0(2):  62. 

Asbtract ( 149 )  

References | Related Articles | Metrics

To improve data availability and performance of cluster, current HDFS adapt uniform data replication. However, different files have different popularity and sometimes

the disparity is enormous, access to high popular data may hurt job performance. To address this problem, a dynamic replicas strategy based on predicted popularity is put

forward. By making full use of the recent data popularity, based on grey prediction model, we use Markov prediction model to correct the predicted deviation because of the burst

access and shifting access, and get the accurate predicted popularity of file. After then, finite channel service model based on the predicted popularity is established to

calculate the minimum replicas meeting user demand. Experimental result shows that compared with default data replication, our strategy can more effectively avoid contentions,

reduce the time consuming of job, and alleviated the network traffic.

Noreference Image Quality Assessment Based on Spatial Natural Scene Statistics

XIA Yujian, SUN Han

2015, 0(2):  67. 

Asbtract ( 214 )  

References | Related Articles | Metrics

In order to measure the image quality of different distortion types, a spatial natural scene statistics (NSS) based model of no reference image quality assessment

method is proposed. The normalized brightness coefficients of natural images obey the generalized Gauss distribution and the proposed method uses these features. The proposed

method can be described as follows. Firstly, the gradient values of natural images are computed in the spatial domain, the ROI (regions of interest) of natural images are

selected by gradient density, the statistical characteristics of natural images are extracted and the multivariate Gauss distribution (MVG) model is established. Secondly, the

same MVG model of test images is established. At last, the quality of test images is evaluated by calculating the distance of the MVG model between test images and natural

images. Experimental results show that the proposed method is more consistent with human subjective perception.

Threedimensional Animation of Basketball Game Based on Unity3D Platform

TANG Yahui1, ZHAO Honglin1, CHANG Yingchun2

2015, 0(2):  73. 

Asbtract ( 262 )  

References | Related Articles | Metrics

Based on Unity3D platform, this article is to model basketball player and venues through 3DS Max and import Unity3D building scenes. Then through animation and games

interactive debugging, it realizes the player’s sports in 3D basketball venues. The system applies the engine system, the physical system, the particle systems and the sound

systems. In order to show the current state of the game, we make the interface display function of system by NGUI plugin, and use Adobe Soundbooth CS4 to handle basketball

sound clips. Thus we can finish the simulation of environment and action of the basketball player when he is doing shoot practice. Then the player scores and game length are

recorded by scoring and timing system, allowing users to feel the shooting sports visually and acoustically.

Design and Implementation of 3D Navigation System for Urban Indoor

WANG Xingfeng

2015, 0(2):  77. 

Asbtract ( 140 )  

References | Related Articles | Metrics

This paper demonstrates a 3D navigation system of urban indoor space guidance. One critical problem of existing guidance systems for the indoor space is that they

mostly deal with twodimensional maps. Moreover, current systems are also not very intuitionist and thus bring troubles in understanding. In this paper, a novel indoor space

navigation system is put forward, which can provide helpful information to guide people to travel in the underground intelligently and conveniently. Different from most previous

systems, it adopts virtual reality simulation technology that cooperates with dynamic display technology, computing technology and communication networks. Through this approach,

users can find and be guided to their destination quickly. The results of simulation show this navigation system is not only feasible but also has effective performance.

Noise Image Edge Detection Algorithm Based on Wavelet Transform

HAO Hongzhuan, ZHANG Weiqiang

2015, 0(2):  80. 

Asbtract ( 164 )  

References | Related Articles | Metrics

Focusing on the problem of noise removal is not complete and edge position is not accurate, this paper presents an improved algorithm for detecting noisy image edge.

The algorithm first improved the gradient algorithm at the time of detecting the edge used by the wavelet transform modulus maxima according to the norm’s nature. Then we used

the improved threshold method which was proposed in this paper to the edge image, and after connecting the image, it obtained the final image edge. Through computer simulation

to validate the algorithm, experimental results show that the algorithm can get the ideal edge image information in noise image, and achieve the bidirectional balance of de

noising and edge precise positioning. The detection algorithm is superior to the traditional method, and it also improves in the aspect of detection effect as well.

A Secure Communication Scheme Based on Multipath Key Agreement for Internet of Things

XIE Wanjuan

2015, 0(2):  86. 

Asbtract ( 139 )  

References | Related Articles | Metrics

Due to the unique characteristics in IoT, such as heterogeneous and processing capacity limited terminals, complex information transmission process and diverse applications, communication security is much more complicated compared with the Internet. Moreover, the existing singlepath transportation security methods can not ensure safety and reduce computation complexity at the same time, which makes these methods not suitable for the IoT. In view of the above issues, this thesis proposes an endtoend secure communication method for the IoT, combining encryption and authentication technology with multipath transportation. The basic idea was as follow: both terminals carried on key agreement scheme based on the DiffieHellman algorithm. Then, the sender used the session key obtained in the above stage and the symmetric encryption algorithm to encrypt data and split the encrypted data into segments, which were then sent to the receiver through multiple independent paths. Theoretical analysis shows that this method can be better against security threats such as maninthemiddle attack, key crack, with reduced computing complexity to a certain extent compared with the singlepath transportation secure method.

An Optimized KFCM Algorithm in Intrusion Detection Based on MCQPSOSA

REN Changrong

2015, 0(2):  90. 

Asbtract ( 194 )  

References | Related Articles | Metrics

Like the classic FCM clustering algorithm and its derived algorithm, KFCM clustering algorithm is sensitive to the initial center and noise data and easy to fall into local optimal value. To solve these problems, this paper proposes a modified fuzzy clustering algorithm based on multipopulations cooperative quantum particle swarm hybrid simulated annealing algorithm(MCQPSOSA). In the improved KFCM algorithm, MCQPSOSA algorithm is introduced to improve search efficiency and global search capabilities. The improved algorithm is used to build intrusion detection system. Our experimental results show the proposed algorithm has more efficient performance which solve poor stability and low detection accuracy of the traditional clustering algorithms in intrusion detection.

A Study of Adaptive Automated Testing to Processes

WANG You1， FENG Shuming1， HE Jinling2， FANG Quan1

2015, 0(2):  95. 

Asbtract ( 144 )  

References | Related Articles | Metrics

It is important to use automated testing techniques in ERP software development for less cost and better repeatability. But, most these techniques can only test one

object, and unable to test the processes of ERP software which may contains many different test objects. And, the correctness of processes of business software is essential.

This paper describes a selfadjusted and automated testing system to ensure the correctness of processes for ERP software. The system uses script database to manage the kinds

of testing scripts, changes each testing script to have the same data interfaces, and maintains the relationships among scripts and processes. Then, the system is able to manage

the scripts in a uniform way, and the scripts can exchange data automatically, and the scripts can be organized automatically to test the changed processes. This paper will use

real example to explain this system.

A Performance Optimization Schema for Huge Load Sakai System

LUO Guizhang1，2，3

2015, 0(2):  101. 

Asbtract ( 197 )  

References | Related Articles | Metrics

To address the problem that server failed and QoS declined in Sakai system when a large number of users participate, a Linuxbased high availability cluster solutions

is proposed. This schema achieved high loadbalancing by HAProxy, Hotstandby mechanisms as ActiveStandby and ActiveActive were built for servers. A masterslave, read

write separated MySQL architecture was proposed, and it achieved high concurrent read database service. Experiments show that this schema has good effects in load balancing and

database performance.

Design of Main Control System of Wind Turbine Based on DSP Processor

LI Kunliang， YU Zirong， CHEN Lijuan

2015, 0(2):  106. 

Asbtract ( 150 )  

References | Related Articles | Metrics

The development and research status of wind turbine control system are introduced in this article. Design of main control system of wind turbine is introduced in

detail and project of main control system based on DSP is proposed. TMS320C6713 is used as core processor in this project and hardware module of main control system is built

through design of modules of peripheral sampling and processing. Communication scheme of main control system is built through the adoption of Ethernet and CAN bus communication

and PC software of the system is written through the platform of VS2008 to achieve the remote control of the system. The experiment shows that the system runs stably and

interfaces of users satisfy the demand of the wind turbine operation.

Recognition and Implementation of Cylinder QR Code

LIU Fengjie， CAI Ming

2015, 0(2):  110. 

Asbtract ( 313 )  

References | Related Articles | Metrics

Traditional QR code recognition algorithm is usually applied to the  barcode printed on flat surface only, the barcode on cylinder, because of the distortion of 2D and

3D space, the recognition rate is low. The paper proposes a new method by adding auxiliary rectangles which is parallel to QR code outer contour for 2D space correction when the

background is relatively clear. After getting the 2D correction image, using area reduction for 3D space correction, it is implemented on the Android platform. The result shows

this method has high recognition rate.

Demand Forecast of Aircraft Engine

GUO Feng1, BAI Song2, ZHOU Bin1, WANG Yan3, LI Yongtao2

2015, 0(2):  113. 

Asbtract ( 130 )  

References | Related Articles | Metrics

The price of aircraft engine was high, and its faults were many, security personnel still adopted experiences to forecast demand, which leaded to the lack of engine or

excessive backlog inventory occurring frequently. This paper puts forward a demand forecast model of aircraft engine by analyzing the demand caused by the engine to life and

random faults, through both ways of fault rate and frequency and unreliability to compute the demand caused by engine to life by aircraft engine’s specified life, remaining

life, single flight task and so on, and the demand caused by random faults. The example shows that the demand forecast model of aircraft engine is some accurate and practical.

A Water Conservancy Metadata Searching and Sharing Platform Based on ElasticSearch

JIANG Kang， FENG Jun， TANG Zhixian， WANG Chao

2015, 0(2):  117. 

Asbtract ( 194 )  

References | Related Articles | Metrics

With the development of water conservancy industry informatization, the massive, multisource and heterogeneous data sharing and discovery become a hot research area.

This paper presents the design and implementation of a metadata search platform based on ElasticSearch, puts forward a solution of water conservancy heterogeneous data and

builds the index on massive data. The multitenant access control policy is used to ensure the consistency and security of user data. The Rest service encapsulate index to

provide searching and multigranularity sharing. Application shows that the platform can ensure users accurately and efficiently to obtain the water conservancy industry data,

save the water conservancy construction cost of unit search system.

Distributed Semantic Retrieval Method for Water Information Resources Directory Services

FENG Jun， XU Xiao， TANG Zhixian， BIAN Yilu

2015, 0(2):  122. 

Asbtract ( 154 )  

References | Related Articles | Metrics

Addressing the demand of high recall rate and realtime for the resource discovery services in water information resources directory services, a distributed metadata

retrieval method (DSRM) based on semantic extension is proposed. It constructs the domain ontology with “Official Document Thesaurus of Water Resources”, achieves the

extension of the specialized vocabulary and the common vocabulary by combining with the semanteme based on Hownet, defines the semantic inference rules and the vocabulary

correlativity, and supports the extension of the query vocabulary by combining with the inference machine. Meanwhile, DSRM defines the threshold of similarity and the selection

method, which can prevent the “semantic drift”, to ensure the searching accuracy. The searching results are ranked according to the combination of the semantic similarity and

the text similarity. The MapReducebased parallelization reform of the index creation and the query processing improves the processing efficiency of retrieval.