International Conference on Applications and Techniques in Cyber Security and Intelligence

In order to ensure driving safety and advanced driver assistance systems (ADAS) attracted more and more attention. Lane departure warning system is an important part of the system. Fast and stable lane detection is a prerequisite for Lane detection under complex background. In this paper, we propose a new lane detection method through a bird’s eye view maps and modified RANSAC (random sampling) based on inspiration from the road feature extraction algorithm for remote sensing images. According to the image of a bird’s eye view, we can identify the tag line through progressive probabilistic Hough transform in the opposite lane detection. Then the group rows are detected by a new weighting scheme based on distance, we can get a candidate lane field. Each field, Lane the RANSAC algorithm is improved and the dual-model fitting. Therefore, the curvature of the road direction can be predicted and the slope of the line. Finally, our results show that lane detection algorithm is robust and real-time performance in a variety of road conditions.

Most face recognition and for monitoring and human – human-computer interaction (HCI) technology tracking system relies on the assumption that in the face of the positive view. Alternative method, the image of face angle to the direction of knowledge can improve performance based on non-frontal view technology. Human face location detection in the city plays an important role in the continuous application of the surveillance video, such as face recognition, face recognition, face a snapshot image screening to save storage capacity. In this chapter, we propose a method for human face location based on Haar features and LVQ technology. First, we performed the eye location based on Haar features. Then, we face image into binary image a number of maps and statistical information on the position of the eyes. After obtaining the statistical distribution of pixels, we based on LVQ neural network classifier classifies the face direction. Based on the results, our algorithm can detect up to 95%. Through the implementation of face direction, we can get the best upright frontal face image recognition and most distinctive quality for further application.

With the development of cloud computing, outsourcing storage-more and more people of all ages. Powerful cloud Server provides customers with a great deal of storage space. In order to protect privacy of outsourcing information, customer information must be encrypted, and redaction. However, it is difficult to search for information in the encrypted file. In this paper, we study the encrypted image retrieval technique of outsourcing. An image retrieval based on content encryption scheme is proposed. In this scenario, using blind method based on discrete logarithm problem of eigenvectors to maintain confidentiality and a clever retrieval method is used. This is a dynamic programme, support a fuzzy search. Clients can control the search. During the search, the original image does not leak.

Cloud platform provides storage space and access for large data sources, cloud computing technology is the key to supporting data technology. However the emergence of large data poses new challenges to traditional data security. This project mainly focuses on data security in the cloud computing environment outsourcing and retrieval, the key technology for solving the outsourcing of data processing, saving local computing resources, reduce computational overhead is of great importance, and for the promotion of health, cloud computing fast and long-term development is of great significance. Therefore, the research is not only of great theoretical significance, and has great practical value.

In recent years, the rapid development of Internet of Things have led to the explosive growth of traffic data. Big traffic data has rapidly developed into a hot topic that attracts extensive attention from academia, industry and governments. The efficient approach to find accompany vehicle is a kind of practices for police criminal investigation department with regard to massive vehicle data retrieval. In this paper, we propose a MapReduce-based approach to find accompany vehicle which contains two MapReduce jobs: the first is to extract the accompany vehicle pairs by traffic monitor position; and the second is to calculate the total frequency of each accompany vehicle pair based on the output of the first job.

Analysis of road traffic accidents is the key to enhance the level of traffic management, big data appears as an innovative technical method. This paper presents a road traffic accident analysis platform based on big data technology. It discusses the key technologies, including content data processing, security, and data analysis. Improvement on public security traffic management can be expected.

The public information platform for logistics is always quite large but not powerful, largely due to the deviation to its operational thought and attributes of its Internet platform. By analyzing the status quo of development of the public information platform for logistics, this paper explores its significance of development. In addition, for the current common problems of the platform, this paper proposes related countermeasures and suggestions on operating the public information platform for logistics with Internet thinking.

In convolution neural networks, digital multiplication operation is the arithmetic operation of the most space-consuming and power consumption. This paper trains convolutional neural network with three different data formats (float point, fixed point and dynamic fixed point) on two different datasets (MNIST, CIFAR-10). For each data set and each data format, the paper assesses the impact of the multiplication accuracy to the error rate at the end of the training. The results show that the network error rate which is trained with low accuracy fixed point has small difference with the network training error rate which is trained with floating point, and this phenomenon shows that the use of low precision can fully meet the training requirements in the process of training the network.

Convolution neural network has important applications in the field of image recognition and retrieval, face recognition and object detection in deep learning. In the training of convolution neural network, 2D convolution, spatial pooling, linear mapping and other operations of forward propagation will have a huge computational complexity. In this paper, an effective optimization technique is proposed to map the convolutional neural network to the digital processor DSP. These technologies include: fixed-point conversion, data reorganization, weight deployment and LUT (look-up table). These technologies enable us to optimize the use of resources on the C66x DSP. The experiment is carried out on Texas Instruments C6678 development board, and the optimization technique proposed in this paper can be applied to multiple open-source network topologies.

With the rapid growth of digital information, it brings a lot of storage burden for resource-constrained users. The powerful cloud server provides huge storage space for users, and it solves the storage issue of huge data. However, the cloud server is not fully trusted, the data has to be encrypted when upload to the cloud server. Keyword search techniques are widely used for the retrieval of encrypted data. Keywords accurate search can be easily achieved. However, fuzzy keywords search is difficulty. In this paper, a fuzzy keyword search scheme is proposed based on comparable encryption technique. In the proposed scheme is flexible, the users can set the similarity of the keywords so as to control the search range.

With the development of network communication and security authentication technologies, Internet finance, a new financial business model which allows customers to achieve online financing, payment, investment and lending, becomes more and more popular. Risk of internet finance is much higher than that of traditional financial system because of the rapid fund flowing, lack of personal credit audit, deficiency of standard network operation and imperfect of information security. Current risk control of Internet finance mainly contains network security, financial self-discipline, investor education and governmental regulatory. In this paper, we propose a risk evaluation method based on structure mining for financial website after analyzing a large number of Internet financial sites. The kernel functions algorithm of natural language syntax tree is introduced to classify the security level of website. While the URL is considered as a long sentence and the path segments are defined as keywords. The experimental results demonstrate that the structure mining method can simply evaluate the risk of Internet financial website to achieve acceptable accuracy.

Word vector and topic model can help retrieve information semantically to some extent. However, there are still many problems. (1) Antonyms share high similarity when clustering with word vectors. (2) Number of all kinds of name entities, such as person name, location name, and organization name is infinite while the number of one specific name entity in corpus is limited. As the result, the vectors for these name entities are not fully trained. In order to overcome above problems, this paper proposes a word vector computation model based on implicit expression. Words with the same meaning are implicitly expression based on dictionary and part of speech. With the implicit expression, the sparsity of corpus is reduced, and word vectors are trained deeper.

Remote sensing has the characteristics of multi-temporal, multi-semantic and multi-spectral, and it plays an important role in most kinds of fields, so we should take some measures to protect the security of the remote sensing. With the rapid development of technology in three-dimensional remote sensing, it has promoted the remote sensing data growth explosively, and it has shown the big data characteristics obviously. Because of the non-complete trusted cloud environment, we consider a security homomorphic encryption scheme over the most significant bit in cloud, which can support the operation for ciphertext remote image in cloud.

Genetic Algorithm is an intelligent algorithm for simulation of biological evolution, is widely applied to solve all kinds of problems. In this paper, several Frequently-used selection operators of Genetic Algorithm are programmed by C language, and are tested in an optimization problem.

With the rapid development of information technology, many kinds of personalized information services have come forward. A key issue is how to represent knowledge effectively, which has a great impact on the personalized service quality. Existing approaches seem to lack the cognition characteristics, which makes information services unable to meet the users’ current cognition level. This paper proposes a novel approach of domain knowledge representation based on the specific academic application background, which shows not only the academic concepts in a specific research filed, but also the logic relation between them. It makes the knowledge representation contains abundant semantics. And we propose the concept cognition energy to evaluate the contribution and value of concept to the specific academic domain, which enhances concept’s domain correlation. Furthermore, the approach presents a hierarchical structure according to the concepts’ profession degree in the field, which ensures the knowledge representation to have the characteristic of cognition. Experimental results demonstrate the effectiveness of the method.

The accuracy of textual keyword extraction is a major factor which influences the text semantic processing. Up to now, there is still much room to improve the precision of textual keyword extraction. To solve the problem, this paper proposes a method to optimize the textual keyword using priori knowledge. First, some priori knowledge for keyword extraction is discussed. Then, a keyword quality evaluation method based on semantic distance between keywords is proposed to judge whether a keyword is good or bad. Next, a textual keyword optimization method is proposed based on the keyword evaluation. Finally, some experiments are carried out, the results of which show that the proposed method can improve the accuracy of keyword extraction on domain texts.

In order to reduce traffic accidents and road congestion in many cities, vehicle speed estimation is very critical and important to observe speed limitation law and traffic conditions. In this paper, we present a speed estimation method of vehicles based on road monitoring video-images. Firstly, we set up a word coordinate system on the license plate in one vehicle image. Next, for small vehicles in China, according to the known length and width of the license plate, the spatial transformation matrix between the word coordinate system and the image coordinate system is derived. Then, based on the spatial transform matrix, compute the corner spatial coordinates of the license plate in each frame, and then estimate the vehicle speed. Finally, experimental results on read data have shown that the vehicle speed can be estimated within the acceptable error range (±3 km/h), and then have demonstrated the effectiveness of the proposed vehicle speed estimation method.

Data leakage is a potentially important issue for businesses. Numerous corporate offer data loss prevention (DLP) solutions to monitor information flow, and detect such leakage. Adding a secret label to a document, DLP can use documents label to do securely control, effectively protecting data. With the increasing documents every day, manual labeling is time-consuming. To better solve the difficult task, recently researchers need to start use document security identification by machine learning quickly classify a large number of texts. The contribution of this paper is to explore dimensionality reduction by feature selection and combine two models to avoid the process of weighting different type of features. In contrast to training all features with one algorithm, our experimental results demonstrate that the combination of two models can improve the classification performance.

In the past 20 years, with the continuous growth of the prefabricated component supply chain, the integration of fragmented information in the supply chain has aroused wide attention. At present, the information of all aspects in the supply chain is isolated, and the problem of the separation of each ring is serious, which not only results in the isolated decision-making of the parties and the waste of resources, but also lead to inefficient supply chain. B2B come into being, which provides real-time data and information interaction for the parties in supply chain, and improve the overall efficiency of the supply chain. This paper focuses on the problem of supplier matching, in B2B platform, proposing a collaborative filtering recommendation algorithm based on matching suppliers, which recommend suppliers for the buyers accurately and improve the overall efficiency of the prefabricated construction industry supply chain.

Illumination, occlusion, pose and expression variations are the most common challenging problems for face recognition in many real-world applications. However, existing face recognition methods are proposed to handle part of these variations. In this paper, we propose a robust facial descriptor to address this issue. First, we apply a chain of three processing to tackle the illumination variation. Second, we compute the facial sparse local descriptor to handle the occlusion, pose, and expression variations. Experimental evaluation on the FRGC database shows that our approach is able to achieve very promising recognition rates under uncontrolled environments.

Recently, computer vision based on deep learning is developing rapidly. As an important branch in this area, face recognition has made great progress. The state of art has achieved 99.77% [1] pair-wise verification accuracy on LFW dataset. But the face dataset in the real application environment such as security checking in the station and bank account opening is much more complex than LFW because of face shelter, postures, uneven illumination and the different resolutions and so on. Except that, LFW dataset only contains the faces like western people but little of other area. Since faces from different areas have not consistent distribution, their methods always cannot achieve high recognition accuracy in practice. In this paper, aiming at Asian face, we propose a multiple-step model training method based on CNN network for real scene face recognition in the absence of large amounts of appropriate data. In the whole training process, each step plays an important role. For step1, it mainly enhanced the generalization ability of model by using a large-scale data set from different source. For step2, it improved the specificity of the model by using a smaller dataset which has closer data distribution in the real scene. And for the final step, metric learning is used to make the model more discriminative and expressive. Meanwhile, some strategy including data cleaning, data augmented and data balance are used in our method to improve the whole performance. Experiments show that this method can achieve high-performance for face recognition in the real application scene.

Fourth times a million police officers held in October 21, 2016 at the Central Political Committee learning seminars, the Political Bureau of the CPC Central Committee and the central politics and Law Committee Secretary Meng Jianzhu pointed out, we are in the era of big data, modern science and technology in the mobile Internet, big data, cloud computing and artificial intelligence as the representative is changing our mode of life, everything experience of human life are changing. Big data development of the human “third eyes”, through massive data analysis, processing, mining, allows us to penetrate into the unknown world. To cultivate data culture, good at using big data thinking analysis, problem solving, decision support.

With the rapid development and wide application of big data technology, a huge amount of data is gathered into big data platform, not only from a wide variety, but also with rapid growth speed. While improving social economic and making social benefits, big data technology is facing great risks and challenges in the aspect of big data security and privacy. Currently, big data privacy has become an urgent problem in the era of big data application which attracts a large number of reports and concerns, and its importance and urgency can’t be ignored. This paper first describes the characteristics and categories of big data privacy, then analysis privacy risks during the whole life cycle of big data processing in deep, including data collection, data integration and fusion, data analysis and data sharing, etc. Finally, this paper discusses the goals and solutions on how to control and prevent big data privacy risks.

With wide application of deep learning in security field, using it on vehicle brand, style and years recognition product has become an active research. Due to the variety of vehicle brand, the total quantity of training samples needed by deep learning is so huge that the difficulty of sample collection and corresponding cost on time and labor are both unacceptable. In addition, new vehicle types come out continuously which require database augmentation and product update in time. To solve this problem, this article proposes a vehicle model data classification algorithm based on hierarchy clustering. Firstly, train the classification model with vehicle data collected by the index of vehicle model information. Secondly, get mean feature of each class and use hierarchical clustering according to the distance between the classes. Then on the basis of distance sorting and model test result to merge the vehicle models. Finally, the feasibility of this algorithm is verified through the experiment. Experimental results show the scheme is feasible. The algorithm realizes the automatic clustering of vehicle model data whose car face or tail has the same structure which can’t be distinguish in image or video. This article provides a new way for the development of vehicle brand, style and year recognition products.

The paper introduces the research on smart companies at home and abroad in combination with smart company characteristics to give the collaborative innovation system architecture. The applications in both Media Group and Sany Heavy Industry are taken as examples for arguments, so as to verify the superiority of the collaborative innovation system architecture.

The emergence of the Internet of things has its specific historical background, the economic weakness of western developed countries and the growing pains of emerging developing countries for the formation of the Internet of things created internal demand. This article predicts the size of the Internet of things market of China in the next six years by using the grey forecasting model and then analyzes the countermeasures for the development of China’s Internet of things market.

This paper presents a simple construction of an efficient publicly verifiable secret sharing scheme (PVSS) in hierarchical settings that uses bilinear pairing (BLP) maps. Till date, hierarchical secret sharing was confined to public key infrastructure (PKI) settings. Use of BLP maps in our scheme yields better security and verifiability. Communications between the Dealer and participants is achieved using an efficient certificateless signcryption (CLSC) scheme. Comparative study with prominent schemes exhibits superior performance of our scheme.

Smart grids are considered as the next generation power grids instead of the traditional power grids. Smart grids provide more efficient power management, more accurate electricity distribution and more reasonable billing statistics. With the deployment of smart grids, security and privacy issues have aroused more and more concern. In this paper, we propose a new privacy-preserving data aggregation scheme in smart grids, which enables a gateway (acted as an aggregator) to aggregate the electricity usage data of users in two dimensions. The new scheme also supports the fault-tolerant property and only needs a little communication by the smart meters.

We aim at creating ease in authentication process through non-password-based authentication scheme for the Dementia patients. The chronic neuro-degenerative disease leaves the patients with memory recall/loss issues. With ever growing rich list of assistive technologies, that bring ease in patient’s daily life i.e. remote Electrocardiography and peripheral capillary oxygen saturation monitoring, remote blood glucose level monitoring applications etc. These assistive technologies are ubiquitous, seamless, immersed in the background, often remotely monitored, and the most intimate applications that run very close to the patient’s physiology. In this paper, we investigate the existing technologies and discover the trends to build Yet Another Authentication Method (YAAM). The YAAM is going to extract a distinctive image from a patient’s viewfinder and securely transform it into authentication token that are supported by the Geo-location, relative proximity of surrounding smart objects etc. that we call security-context. The authentication tokens are only generated on the fly when token context is right for the image stream captured by the wearable camera. The results presented in this paper not only present the pros and cons of the existing alternative authentication technologies, they also aide in the development of the YAAM prototype.

The order of input is an important reason for a fault to take place. Most specifically, in the even driven software where multiple events run one after another and action of one event depends on another one. In such a system, a fault is usually identified on a state when some events have already been occurred. To identify this fault, a sequence covering array is created ensuring that a sequence of a required t-way or pairwise (interaction) events are covered. However, generation of optimum sequences appeared to be a NP-hard problem. In the paper, we adopted swarm intelligence to generate the sequence covering array and a novel technique known as SISEQ is proposed. In the end, the SISEQ is compared with other technique. Finally, the analysis section shows that our technique is more acceptable.

The limitation of resources and the deadline of software and hardware projects inhibits the exhaustive testing of a system. The most effective way to overcome this problem is to generation of optimal test suite. Heuristic searches are used to optimize the test suite since 1992. Recently, the interest and activities is increasing in this area. In theory, the changes to the parameter interaction (the t) can significantly reduce the number data in the test suite. Using this principle many scientists and practitioners created some effective test suite generation strategies. The implementation of heuristic search in the generation of optimum and minimum test suite is the most effective. However, producing the optimum test data is a NP-hard problem (Non-deterministic polynomial). Thus, it is impossible for any strategy that can produce the optimum test suite in any circumstance. This paper represents a novel swarm intelligent based searching strategy (mSITG) to generate optimum test suite. The performances of the mSITG are analyzed and compared with other well-known strategies. Empirical result shows that the proposed strategy is highly acceptable in terms of the test data size.

With the emergence and extensive deployment of biometric-based user authentication system, ensuring the security of biometric template is becoming a growing concern in the research community. One approach to securing template is to transform the original biometric features into a non-invertible form and to use it for a person’s authentication. Registration-based template protection schemes require an accurate alignment of the enrolled and the query images, which is very difficult to achieve. To overcome the alignment issue, registration-free template protection approaches have been proposed that rely on local features such as minutiae details in a fingerprint image. In this paper, we develop an alignment-free fingerprint template protection technique which extracts the rotation and translation invariant features from the neighbouring region of each minutia and then exploits the neighbourhood information to achieve the non-invertible property. Evaluation of the proposed scheme on FVC2002 DB1-B shows that the new method exhibits satisfactory performance in terms of recognition accuracy, computational complexity, and security.

Exfiltration of sensitive data by malicious software or malware is a serious cyber threat around the world that has catastrophic effect on businesses, research organizations, national intelligence, as well as individuals. Thousands of cyber criminals attempt every day to attack computer systems by employing malicious software with an intention to breach crucial data, damage or manipulate data, or to make illegal financial transfers. Protection of this data is therefore, a critical concern in the research community. This manuscript aims to propose a comprehensive framework to classify and detect malicious software to protect sensitive data against malicious threats using data mining and machine learning classification techniques. In this work, we employ a robust and efficient approach for malware classification and detection by analyzing both signature-based and anomaly-based features. Experimental results confirm the superiority of the proposed approach over other similar methods.

Nowadays, the cameras of traffic monitoring systems are mounted toward roads or interactions. The views are fixed and limited. To extend the monitoring area, a novel onboard abnormal event detection system is proposed based on the In Vehicle Monitoring System (IVMS). Videos are captured by the camera mounted in the front of a vehicle. Traffic information extracted by the system is combined with the GPS and electronic map to discover abnormal events. The system can be installed on police vehicles, buses, even private cars. Therefore, the anomaly monitoring can be processed anywhere, instead of on the certain interactions or roads.

The Code Injection Attack (CIA) exploits a security vulnerability or computer bug that is caused by processing invalid data, CIA is a serious attack problem that attackers try to introduce any new methodologies to bypass the defense system. In this paper, we introduce a novel detection algorithm for detection of code injection attack. Our empirical performance shows that the proposed algorithm give better results compared to existing results.

Many studies have shown the weaknesses in MIFARE Classic, which is the most commonly used in access control systems, and conducted several attacks successfully. But in the situation of multi-section attacks, it would cost long time to retrieve the key of Crypto-1 cipher which is used in MIFARE Classic. We have designed a new algorithm to retrieve the key of Crypto-1 based on parallel computing using GPU so that we can reduce the time consumption for multi-section attacks. We have implemented and optimized our algorithm using CUDA and OpenCL, and tested them on different platforms contrast with the traditional method using multi-core CPU. Experimental results show that our algorithm is quite efficient on a GPU and get better performance than the traditional method on a 12-core CPU. This should be a better method to retrieve the key of Crypto-1 cipher for multi-section attacks.

In recent years, the electronic health record (eHR) system is regarded as one of the biggest developments in healthcare domains. A personally controlled electronic health record (PCEHR) system, offered by the Australian government makes the health system more agile, reliable, and sustainable. Although the existing PCEHR system is proposed to be fully controlled by the patients, however there are ways for healthcare professionals and database/system operators to reveal the records for corruption as system operators are assumed to be trusted by default. Moreover, as a consequence of increased threats to security of electronic health records, an actual need for a strong and effective authentication and access control methods has raised. Furthermore, due to the sensitive nature of eHRs, the most important challenges towards fine-grained, cryptographically implemented access control schemes which guarantee data privacy and reliability, verifying that only authorized people can access the corresponding health records. Moreover, an uninterrupted application of the security principle of electronic data files necessitates encrypted databases. In this paper we concentrates the above limitations together by proposing a robust authentication scheme and a hybrid access control model to enhance the security and privacy of eHRs. Homomorphic encryption technique is applied in storing and working with the eHRs in the proposed cloud-based PCEHR framework. The proposed model ensures the control of both security and privacy of eHRs accumulated in the cloud database.

Low cost ad hoc networks like Wireless Sensor Networks (WSNs) are best suited to gather sensory information. Sensitivity of these classified information leads to the necessity of implementing security protocols during their exchange. Such implementations use cryptosystems that may suit resourceful Internet of Things (IoT) devices; but overburdens tiny sensors. Moreover most protocols assume that an adversary is well versed with all system information, barring the cryptographic keys. As such the (fixed) operational frequency bands between a given pair of nodes is assumed to be known at all times. Such a strong assumption may not be always necessary in real life deployment zones. In fact tracking an operational frequency between sensors from a range of bands may be difficult in a large network [15]; though not hard. This leads to a hard problem, i.e., to keep track of recursive switch of operational frequencies between a given pair of sensors for consecutive timestamps. We exploit hardness of this problem to achieve confidentiality of message exchange between pairs of nodes. Message to be transmitted is split using secret sharing technique [18]. Each piece is then transmitted via different bands obtained by recursive use of cryptographic hash function on initial preallocated bands. Our approach does not consume extra energy during message transmission or receipt in comparison to existing wireless systems. Storage requirement is minimized to storage of hash functions; no cryptographic key stored. Security achieved is comparable to any existing cryptosystem.

In smart grid communications, preserving the privacy of consumers’ electricity usage data is a topic of interest for power providers and consumers, as well as regulators. Sui (IEEE Trans. Smart Grid, 2016) proposed a new threshold-based anonymous identification (TAI) scheme for smart grid communications and claimed that TAI scheme achieves unlinkability, strong anonymity, non-frameability, identification, and integrity. In this paper, however, we demonstrate that due to a flawed Decisional Diffie–Hellman assumption in a bilinear group, TAI scheme is unlikely to achieve unlinkability, in violation of their security claims. Specifically, an adversary $$ {\mathcal{A}} $$A can easily link different consumption reports from the same consumer during the anonymous consumption reporting part and link a disavowal proof of a compliant smart meter to its previous signature. We then propose an enhanced anonymous identification scheme to eliminate the security vulnerability in the scheme, in the sense that no one can determine whether two different consumption reports are from the same consumer.

This paper presents a step-by-step demonstration for the exploitation of CVE-2014-6271, affecting the ‘Bourne Again Shell’ (Bash). By design, Bash cannot be accessed via a web server; yet a flaw in its source code provides attackers the ability of Arbitrary Code Execution (ACE) over a Common Gateway Interface (CGI). In this paper, we demonstrate how Shellshock vulnerability can be exploited, as well as outlining mitigation strategies.

As a compact and efficient way to present relational data information, Web tables are used frequently in Web documents. Web table positioning technology are considered as essential components of Web table information extraction, and more and more people pay attention to them. This paper realizes table positioning according to Web table structure label and heuristic rules of user-definition, which includes the solution of <TABLE> nested problem, the determination of table data’s integrity, and traversal of <TABLE> tree. The experimental results show that our web table positioning method has good performance.

The big data service platform of public security information brings together all kinds of data resources related to public safety, and provides an effective data base for big data application analysis system. However, centralized data resources presents a huge challenge to data security protection services. This paper studies the data security protection mechanism from the three main aspects of data storage, data management and data service, which combines the characteristics of public security business and public security big data platform.

Openstack has become a management cloud computing operating system standard of public cloud, private cloud and hybrid cloud in recent years. In this paper, we detailly describe Openstack architecture and provide an experimental method for using Devstack as automated scripts tool to deploy an Openstack cloud platform in a stand-alone environment. Application characteristic of local, flat, vlan and vxlan four types of tenant Openstack network based on Linux bridge is mainly presented, which is contributed for cloud tenants with their web browser to build the network infrastructure including computers, switches, routers and firewall in a very short period of time.

It is a hot study topic that how to obtain the food collocation and the effect of food collocation. The extraction method for best match of food nutrition is proposed to solve the problem in this paper. First, the method of forward maximum matching is used to segment sentences and filter stop words. Then, the nutrition content of food is abstracted as food collocation vector. At last, the classification results of the KNN algorithm are used to identify the verb of sentence. The average accuracy of the test is 45.9%. The experiments show that the method is effective.

In the traditional discovery methods of micro-blog new login word, compound words are difficult to be extracted effectively. Aiming to solve this problem, this paper proposes an extraction method of micro-blog new login word based on improved Position-Word Probability (PWP) and N-increment algorithm. First, the micro-blog long text is composed of all micro-blog within a single topic in period of a given time and then pre-treated. Then, the extension direction of frequent strings is judged by improved the probability of word location in the query process of N-increment algorithm. Finally, the redundant strings are reduced by pruning frequent strings set. The experimental results show that the algorithm proposed in this paper can effectively extract the compound words in micro-blog new login word.

To improve the access speed and efficiency of excessive and low efficiency micro-blog data, this paper presents a method for building the knowledge flow of micro-blog topic. The core task of building the knowledge flow of micro-blog topic is analyzing each micro-blog information (e.g., the number of point praise, the number of forwarding and the fresh degree for micro-blog) to realize the organization of micro-blog topic. First, we collect and process micro-blog information, including the number of point praise, the number of forwarding and the fresh degree. Then, based on the achieved the information of each micro-blog, we do the filtering of micro-blog to keep that interesting/meaningful micro-blog. And we sort all the kept micro-blog messages browsed by a user to generate a knowledge flow of micro-blog topic. The experimental results show that the proposed algorithm has a high accuracy.

At present, more and more data are generated every day and the actual application requirements for the mining algorithm efficiency have become higher. In such a situation, one of the hot research topics on the frequent pattern mining over uncertain data is the spatiotemporal efficiency improvement of mining algorithms. Aiming at solving the frequent pattern mining problems over dynamic uncertain data streams, based on the existing algorithm researches, the paper proposes a parallel mining approximation algorithm based on the MapReduce framework by combining a highly efficient algorithm for static data. If this algorithm is used to mine frequent patterns, all the frequent patterns can be mined from a sliding window by using MapReduce at most twice. In the experiments conducted for this paper, in most cases the frequent item set was accurately discovered after MapReduce is used once. The experiments have shown that the spatiotemporal efficiency of the algorithm proposed in this paper is much better than those of the other algorithms.

The rolling planning of distribution network is the most essential section in the development of smart distribution network. Based on the big data collected from inside and outside of the electric power industry, this paper analyzed the geographical, social and economic overview, planning year and basis for planning, the demand for the rolling planning and the current state of the distribution network and provide more feedbacks for grid operation. Then the load prediction through the growth rate method, linear regression method and comprehensive power consumption per capita method etc. is also discussed in this paper. Finally, according to the predicting results of the power demand forecasting during the planning period, the objective of distribution network rolling planning and prospects of 110 kV distribution network in Longhai are also provided. So it is very significant to rectify and optimize the search on rolling planning of distribution network and thus lay reliable scientific basis for long-range distribution network perspective and construction.

Using the method of experimental research, the experimental group and the control group of 10 female students were tested with electroencephalogram (EEG). The values of concentration and relaxation were collected in three states, i.e. quiet state, 1 min and 3 min at the end of 3-min step test. Based on the comparative analysis, the changes of EEG in aerobic exercise were revealed, and then the EEG model was established. The training program was adjusted at any time on the premise that the model was close to the EEG. After 3 months of endurance training, the results were analyzed and the optimal scheme was selected. The results show that: 1. According to the EEG model established by SPSS MODELER, the exercise effect of experimental group 2 in walking and running training program adjusted anytime is better than that of experimental group 1 of traditional exercise; 2. EEG can be used as an important reference index to guide the development of aerobic exercise and improve endurance quality.

Clustering of XML documents is a useful technique for knowledge discovery in XML databases. However, the process of clustering XML documents is always time-consuming due to the semi-structured characteristics of the documents. In this paper, we present an efficient clustering algorithm called Frequent Edge-based XML Clustering (FEXC) to cluster XML documents using frequent edge sets. First, we represent XML documents using edge sets, and then discover the frequent edge sets for each document employing a traditional frequent pattern mining approach. Second, for each frequent edge set, we find all the documents containing it, and then compute a measure called entropy overlap, which indicates the document relevance (overlap) with the ones containing all other frequent edge sets. Clustering is then performed using the entropy overlap measure. Third, we perform a merging process which removes redundant clusters, therefore reducing the number of clusters. Experimental results show that our proposed method outperforms the traditional distance-based XML clustering algorithm in terms of efficiency without compromising the quality of clustering.

With the continuous expansion of e-commerce applications in China, people not only enjoy the conveniences, but also encounter the difficulties of being unable to find their demands from a large number of e-commerce goods. On the basis of combination of Hadoop distributed system infrastructure with traditional collaborative filtering recommended algorithm, the sales records of the existing tea sales system is analyzed in this paper, so as to obtain the recommended principle that meets consumer preference and help users to find the tea they need more quickly. This helps tea enterprises to extend their marketing channel and improve tea sales.

Recently, sparse subspace learning (SSL) has been widely focused by researchers. SSL methods aim to project samples into a low-dimensional subspace which can well maintain sparse correlations of dataset. However, most SSL methods utilize sparse representation (SR) which constructs sparse correlations without label information. Therefore, labels can’t be fully utilized to improve discriminative abilities of SSL methods. In order to overcome this drawback, this paper proposed a novel method called semi-supervised sparsity preserving projection (SSPP). SSPP first combines label information with SR to construct sparse correlations between samples. Some wrong correlations are avoided due to the employment of labels. Then, in order to further improve discriminative abilities of SSPP, large-margin criterion is adopted. Various experiments show the excellent performance of SSPP.

Comovement widely exists among financial time series. Although sundry researches have been implemented for studying this phenomenon, manual judgments are still one vital measure for investment strategy decisions. To augment manual analysis on comovement of time series, we propose an animated approach for time series data processing and animation creation. Example calculations are carried out on 8 major Forex currency pairs and resulting movies are presented.

In order to cut down the localization accuracy problem of wireless sensor network (WSN), a novel node localization method is proposed with back propagation neural network (BPNN). At first, the calculation of node localization is presented by ranging interval and signal strength, and the parameters are rapid solving base on BPNN. Finally, a simulation experiment is conducted to study the influence key factor with NS2 and MATLAB. The results show that, compared other localization algorithm, this method has good suitability, and it could effectively reduce the localization error.

Based on in-depth analysis of the current situation of big data applications and existing main problems of China’s commodity exchange market, we discuss the necessity and feasibility of accelerating the application of big data in China’s commodity exchange market, put forward the function framework of the application system of the big data in China’s commodity exchange market, point out developing big data application path of China’s commodity market, and put forward the development direction of big data application in China’s commodity exchange market.

Because of the advantage of centroid localization systembased on FPGA&DSP in machine vision measuring system, Centroid location algorithm for multi-objective is proposed which achieves image histogram in FPGA, computes adaptive threshold based on histogram in DSP, marks the objects in FPGA with modified connected component labeling and then calculates the centroid of the objects. Experimental results shows that the results of Multi-object Centroid Localization System is consistent with the actual results, and the system can run 70 fps in maximum, which is much higher than the processing speed with single DSP or computer. Multi-objects centroid localization system can be applied to machine vision measurement system for its accuracy and real-time.

With the development of the Smart Cities, the security has become an urgent necessity. It refers to an urban transformation which, using latest ICT technologies makes cities more efficient. Composed of a growing Internet of things (IoT), cloud computing, big data analysis, mobile Internet via broadband connections, or objects and sensors via low-cost data links, the greatest challenge today is to meaningfully manage such systems in the widespread virus and attacks. Given that these systems will greatly impact the operation of smart city, issues related to privacy and security has come into limelight. Just like the existence of a biological self-defense system in the body plays the important roles, the digital Bio Self-Defense System (BSDS) is designed to protect the security of smart city including four major defenses: ① Digital Skin is responsible for distributed key generation and storage. ② Immunity System improves fraud prevention and anti-attack ability. ③ Self-healing includes instant snapshots, automatic backup, active recovery strategies. ④ Nerve monitory uses multi-information digital nervous system integration, multi-factor cross-validation methods to achieve self-organizing, adaptive, self-defense capabilities. This paper summarizes the key challenges, emerging technology standards, and issues to be watched out for in the context of privacy and security in smart cities. A key observation is that bionic system, model and norm designed can meet high security requirements of smart city to avoid third party abuses.

In this paper, we first present the intuitionistic fuzzy induced generalized ordered weighted averaging distance (IFIGOWAD) operator. The main advantage is that it provides a more complete generalization of the aggregation operators that includes a wide range of situations. We further generalize the IFIGOWAD operator by using quasi-arithmetic means obtaining the intuitionistic fuzzy induced quasi-arithmetic OWAD (Quasi-IFIOWAD) operator and by using hybrid averages forming the intuitionistic fuzzy induced generalized hybrid average distance (IFIGHAD) operator. Then a new approach based on developed operators are introduced for decision making problem. Finally, a numerical example is provided to illustrate the practicality and feasibility of the developed method

In this paper, a new 2-tuple linguistic aggregation operator called 2-tuple linguistic ordered weighted averaging-weighted averaging distance (2TLOWAWAD) operator is presented. Some of its desirable properties and families are further explored. Moreover, by employing the proposed operator, a method for 2-tuple linguistic multi-attribute decision making problem is developed. Finally, an illustrative example concerning information systems security assessment is provided to illustrate the applicability and effectiveness of the proposed method.

Artificial intelligence is the simulation of the information process of human consciousness and thinking. The search algorithm based on artificial intelligence has a good application prospect. In this paper, the shortcomings of traditional search algorithms are analyzed, and the disadvantages can be made up by artificial intelligence. We first analyze the application directions of artificial intelligence in the search algorithm, and then point out the system requirements and the overall design of the search algorithm based on artificial intelligence. It includes the game module, game board module, player module and displayer module. At the same time, we analyze the functions in the above four modules and construct a search algorithm based on artificial intelligence. Finally, the part codes of the search algorithm are given to provide some reference for the relative researchers.