Abstract:The traditional collaborative filtering recommendation algorithm has such problems as data sparseness, cold-start and new users. With the rapid development of social network and e-commerce, how to provide personalized recommendations based on the trust between users and user interest tag is becoming a hot research topic. In this study, we propose a probability matrix factorization model (STUIPMF) by integrating social trust and user interest. First, we excavate implicit trust relationship between users and potential interest label from the perspective of user rating. Then we use the probability matrix factorization model to conduct matrix decomposition of user ratings information, users trust relationship, user interest label information, and further excavate the user characteristics to ease data sparseness. Finally, we make experiments based on the Epinions dataset to verify the proposed method. The results show that the proposed method can to some extent improve the recommendation accuracy, ease cold-start and new user problems. Meanwhile, the proposed STUIPMF approach also has good scalability.

Abstract:Ecological-Niche Factor Analysis (ENFA) is a multivariable approach based on the concept of the ecological niche. But when computing the relevance between variables by covariance, it only handles linear dependencies, while most is nonlinear interaction. Mutual information measures the interdependence between variables and it's not limited to linear relations. ENFA based on mutual information (MIENFA) is presented which uses mutual information as the relevance. Through studies of Bar-headed Goose in Qinghai Lake, compared with the traditional ENFA, the proposed approach changes the specialization vector and improves the accurate rate of habitat suitability prediction.

Abstract:Phase Change Memory (PCM) has become a candidate of future main memories due to its attractive characteristics of non-volatility, high access speed, and low power consumption. Meanwhile, how to efficiently integrate PCM into current memory systems is becoming a hot topic. Generally, there are a number of choices to use PCM as main memory, e.g., to construct PCM-only main memory systems, or to construct DRAM/PCM-based hybrid memory systems. However, the conflict between numerous PCM-related researches and lack of real devices hinders evaluations of PCM-aware algorithms. Therefore, in this paper, we propose a DRAM/PCM-based hybrid memory simulator. The new features of the simulator are manifold. First, it can simulate different DRAM/PCM-based memory systems, including the hierarchical architecture (DRAM as the cache of PCM) and the hybrid architecture (both DRAM and PCM as main memory). Second, it leverages a clock-accurate timing model to emulate accesses on PCM. Third, it offers a hybrid memory allocation interface that can be easily used by programmers. After a description of the simulator framework, we present basic evaluation results and a case study of the simulator, which suggest its feasibility.

Abstract:Blob is a basic form in the gray-scale image, and has important application in image analysis. In this paper, a novel blob form filter (BFF) is proposed to detect the blob target in image. Based on the principle of percentile filtering, the BFF has strong robustness to noise. The experimental results show that BFF cannot only effectively detect the blob region specified size from binary image and gray-scale image with noise, but can also get the blob form very close to the original form.

Abstract:The thesis defense grouping is a common problem in the college management. To ensure the fairness and scientificity, it is necessary to consider some constraints between supervisors and students when grouping. There are two inherent contradictions:the principles of mutual avoidance and uniformity. In this paper, the main issue is to find out an optimal solution that satisfies the two conditions as far as it is possible. Through the establishment of mathematical model, the respondent grouping problem is summarized as the matrix encoding. Then two conflict conditions are consolidated into one objective function. The single objective differential evolution with superior-inferior crossover scheme is adopted to solve this problem. A suitable chromosome representation and fitness function are designed. A series of operations such as mutation, superior-inferior crossover and modification are performed. The optimal solution is obtained when the evolution is terminated. To test the advantages of this method, a general algorithm is designed for comparison with it. The results show that the grouping solution obtained by differential evolution using a superior-inferior crossover scheme is more scientific and feasible than the general algorithm.

Abstract:BWDSP is a 32bit static scalar digital signal processor with VLIW and SIMD features, which is designed for high performance computing. In order to meet the performance requirements of digital high-performance computing, the soul core DSP provides a rich set of complex instructions, and the compiler cannot directly use these complex instructions to improve the compilation performance. Since BWDSP has a wealth of complex type of instructions, and it has high performance demands in the radar digital signal field, the implementation is researched according to the characteristics of BWDSP features based on the traditional open-source Open64 compiler framework to achieve the complex data type and complex operations support operations, and further optimization of complex instruction is realized by identifying a specific type of complex operation of a series of patterns. The experimental results show that the implementation on BWDSP compiler can achieve 5.28 time performance improvement on average.

Abstract:At present, the construction of ocean information has become an important support for national strategic aim of building an ocean power. However, the problems like the lack of unified planning, serious homogenization, isolated information, not enough wisdom in the construction process become increasingly evident. Starting from the idea of the integrated planning and construction, this paper puts forward a scheme of smart ocean engineering construction based on human body theory and the modern information technology of Internet of Things, Big data and artificial intelligence. This scheme completely designs the three-dimensional monitoring system of ocean environment, big data platform of ocean, the ocean application system as well as the information standards and norms, and other aspects. This scheme covers multiple business categories such as the ocean environmental protection, disaster prevention and mitigation, seas and islands, ocean law enforcement, ocean economy, and ocean fisheries. The results of this research are applied in the construction of smart ocean in coastal provinces and countries. Besides, it is of great significance to the construction of smart ocean in other coastal provinces and countries.

Abstract:The current application of big data in universities is facing many problems such as difficulties in data integration, results application and knowledge management, which need to be resolved urgently. Combined with the method of knowledge management and software engineering, this paper firstly designs the process of university big data mining based on CRISP-DM. On this basis, it designs the overall structure of the university knowledge management system and the function models, and then it designs the main function models in detail. Next, it analyzes the key technologies of data integration and big data mining. Finally, it gives the development environment and test of the system. The design scheme of the system provides a reference for the application of big data in universities.

Abstract:As a strategic resource of technological innovation and development of the oceans in China, marine science data should be the element for sustainable development. However, marine scientific data involve many disciplines, complex types, diverse sources, and different formats, and hence it is difficult to process the data and visualization. In terms of the characteristics of marine scientific data, the platform achieves the integration, analysis and visual processing of marine data management by introducing the WebGIS technology, combined with marine scientific data services.

Abstract:With the development of semantic web technology, RDF data grow rapidly. The single node RDF query system cannot meet the practical needs. Building distributed RDF query system has become one of the hotspots in the academia and industry. The existing RDF query system is based on Hadoop and general distributed technology. The disk I/O of the former is too high and the latter is less scalable. Besides, the two systems perform poorly in the basic pattern matching mode. In order to solve these problems, we design a distributed system architecture based on Spark and Redis, and improve the query plan generation algorithm. We call the prototype system RDF-SR. This system reduces the disk I/O by Spark, improves the data mapping rate by Redis and reduces the data shuffling process with improved algorithms. Our evaluation shows that RDF-SR performs better in the basic pattern matching mode compared with other systems.

Abstract:To tackle the problem that the extraction of the curved trajectory is difficult and the operation is complex in manual teaching, this paper uses the way that Visual Studio makes the secondary exploitation for the SolidWorks software to make the off-line programming of the curved surface of robot, and loads it automatically into SolidWorks in the form of DLL plug-in. By using UV parametric curved surface, the system realizes the rapid extraction of off-line trajectory and the generation of robot control code and finally implements the motion simulation of the robot with the model of the robot showed by DH parameter by means of traversing the joint angle. Practice has proven that the control code generated by the off-line programming can be directly downloaded to the robot controller which reaches the expected effect. Such a system is of great significance in the actual development and application of the industrial situations and domestic small and medium-sized enterprises for the rapid machining various parts.

Abstract:Xiangshan Culture has multiple elements and rich contents, but the isolation and the scattering of each research is due to the lack of systematic information organization. After summarizing the applications of knowledge graph information organization, this paper provides a method of Xiangshan Culture information organization based on knowledge graph. By using the Bone-method, the knowledge graph of Xiangshan Culture has been built based on ontology, with the connections among the scattered research topics of Xiangshan Culture. The knowledge graph-based Xiangshan Culture retrieval system visualizes complex knowledge instances and knowledge relations, in order to sketch the overall contours of Xiangshan Culture. The information organization based on knowledge graph has the advantage of analyzing and digging complex domain relations, while the knowledge graph retrieval system has vivid visualization models and distinctive features.

Abstract:A real-time message pushing system, including teacher side and student side, is designed and implemented based on the WebSocket protocol aimed to improve the low efficiency in information exchange between teachers and students and to overcome the difficulty in getting the statistical feedback. At the same time, the key models and major technology are thoroughly investigated and analyzed. The results of the test indicate that the proposed system is stable and robust, and it meets the needs of message pushing software in terms of high efficiency and property in real-time.

Abstract:ETL(Extract-Transform-Load) is an indispensable component of the NC machine sensor data analysis. In view of the increasing amount of data in recent years, and the rising requirements for real-time data processing, this paper improves the traditional change data capture methods, and designs real-time data capture solutions based on the characteristics of those data. What's more, it details the entire ETL distributed architecture design. Finally, experiments show that the ETL system on NC machine sensor data real-time processing has higher efficiency.

Abstract:The application software under the traditional java framework lacks the ability of modularization and dynamic management. Based on the in-depth study of OSGI framework, this paper proposes a method of combining the OSGI and the hierarchical decoupling. It uses the method to design the overall architecture of downhole operation management system, and also solves the problem of dynamic management and service layer. It describes the application of OSGI extension point mechanism and AOP in detail, then improves the system expansion ability and solves the problem of data synchronization update. The test results show that the model can improve the performance of the system, reduces the coupling between the modules, increases the reusability of components and scalability, and improves the stability of the system.

Abstract:In event-based social networks (EBSN), groups that aggregate users with similar interests for sharing events play important roles in community development. Understanding why people join a group and how groups are formed is particularly an interesting issue in social science. In this paper, we study predicting users' preferences on social groups by considering content information in EBSN, i.e., geographic-social event-based recommendation. Specifically, we consider two types of content information, i.e., the tags and geographical event locations about users/events. We propose the SEGELER (pair-wiSE Geo-social Event-based LatEnt factoR) to model the users behavior considering the information. Experiments on a real-world EBSN social network validate the effectiveness of our proposed approach for both normal users and cold start users.

Abstract:With the rapid development of information technology, the number of electronic products shows a great increment. Personal data resources are distributed in various devices and therefore efficient data sharing among devices becomes a problem. Given the fact that the existing methods can not meet the demand of efficient data sharing in this circumstance, this paper proposes a personal data resources sharing method based on service-oriented technology. It is an end-to-end method of sharing data that every end runs three basic modules, which are data web service module, service publication module and service discovery module. Compared with the existing data sharing scheme, it has advantages of avoiding additional hardware maintenance, high sharing efficiency in Local Area Network, supporting online access, no need to worry about the data leakage by third parties and a low system resource occupancy rate. The method is an effective solution in the data sharing environment to meet the challenges of frequent changes, information consistency maintenance, heterogeneous nodes and acceptable cost.

Abstract:Color cast detection is widely used in machine vision and related fields. Traditional algorithms mostly depend on certain assumptions or prior knowledge. In this paper, a new cast detection approach based on color-feature analysis is proposed, which involves image colorfulness and color naturalness. Experiments show that this approach can eliminate the influence of dominant colors, improve the detection accuracy, match the subjective evaluation according to the human visual system, and is effective on more occasions.

Abstract:In order to improve the search ability of the cuckoo search algorithm, this paper proposes a quantum-inspired cuckoo search algorithm by introducing the quantum computing mechanism into the classical cuckoo search algorithm.. In the proposed algorithm, the qubits are used to encode individuals, and the Pauli matrixes are employed to determine rotation axis. The Levy flight principle is applied to obtain rotation angle, and the rotation of the qubits on the Bloch sphere is used to update the individuals. The experimental results of extreme optimization of benchmark test functions show that the proposed algorithm is obviously superior to the classical cuckoo search algorithm in optimization ability.

Abstract:Aiming at the problem of interoperability between two information systems, a formalized method about interoperation research based on polychromatic sets theory is proposed, and the research steps of interoperability between two systems based on polychromatic sets theory are given. Firstly, the polychromatic sets theory is introduced and its contour is segmented to reflect the static and dynamic properties of the objects. The concept of single element polychromatic sets and its aggregation operation are given. Secondly, according to polychromatic set theory and its extensions, the research steps of interoperability between two information systems are given. The interoperability of the two systems are described in a single element polychromatic sets, and then the whole entity of interoperation is reflected through the union operation of two single element polychromatic sets. Finally, the validity of this formalized method is verified by the interoperability between hospital information system (HIS) and laboratory information system (LIS).

Abstract:With the rapid development of Internet, the traditional personalized recommendation involving only users and projects cannot meet demands in efficiency and accuracy of the recommendation. Therefore, context-aware recommendation has drawn wide attention and become a new research hotspot. This paper analyzes the definition of context and the model of context-aware recommendation. It also proposes an association rule recommendation model based on context information which reduces the number of dimensions. The data source of experiments is web log. Finally, this paper combines the temporal context and implements the association rule recommendation algorithm based on the temporal context partition. Compared with the traditional recommendation algorithm, the results of experiments show that the context-aware recommendation algorithm has higher accuracy and recall rate.

Abstract:When the computer is handling multi-file tasks, it may read and write a file at the same time, resulting in the failure of the file data to be fully read and written or in the loss of some data. In the Linux kernel, with the single processor, task allocation and processing is made with the synchronization mechanism. The classic approach is atomic operations, semaphore mechanisms, mutexes, etc. In the multi-processor systems, the test-and-set primitive operation is made to solve the problem. In this paper, we design a new task scheduling scheme for multitasking to avoid mutual exclusion access. We use a Matlab program to realize the algorithm, and the result shows that the algorithm can effectively realize the multi-file tasks parallel execution.

Abstract:The correct use of collocation has been widely acknowledged as an essential characteristic to distinguish native English speakers from English learners. Through the analysis of CLEC, we can find that English learners often make mistakes on verb-noun collocations. In this paper, we propose a hierarchical language model that can be used to correct verb-noun collocation errors made by English learners. The language model takes the dependencies between words within a sentence into account. It parses sentences into different levels of clauses. The words within the same clause are highly correlated, and the relevance of words in different clauses is weak. The language model is more stable. Moreover, it is more accurate because collocation information is condensed. It can be used to re-rank candidates and generate classifier features. We apply this hierarchical language model to the correction of English verb-noun collocation errors. Compared with the traditional language model, the new model has better performance.

Abstract:Every technological advance in the development of human society will lead to a series of new products and services, but at the same time, it will lead to a sharp increase in the consumption of resources and energy. Although the progress of technology has improved the efficiency of resource and energy use, the development mode of increasing per capita energy consumption is not sustainable. In addition to the supply side of energy efficiency, a reasonable guide to demand side can be the key used to achieve building energy efficiency. To achieve the change of building energy-saving mode from the supply side to the demand side, we must properly describe the energy by using characteristics in a specific indoor environment, and assess the reasonableness of building energy consumption from the demand side, then identify accurately the reason of energy waste. Along with the rapid development of building automation system and IOT technology, a large amount of building energy consumption data with specific indoor environment features are acquired, then we can use data mining technology to extract energy saving clues and strategies from these low density value building daily operation data. This paper briefly introduces the data mining technology, and summarizes the application of various mining methods in building energy saving, and prospects of its development trend.

Abstract:In the paper we explore the two important parts of music emotion classification:feature selection and classifier. In terms of feature selection, single feature cannot fully present music emotions in the traditional algorithm, which, however, can be solved by the multi-feature fusion put forward in this paper. Specifically, the sound characteristics and prosodic features are combined as a symbol to express music emotion. In the classifier selection, the deep belief networks are adopted to train and classify music emotions, which had a better performance in the area of audio retrieval. The results show that the algorithm performs better than the single feature classification and SVM classification in music emotion classification.

Abstract:The positioning system based on WIFI location fingerprint can achieve high precision indoor location. The neighbor selection algorithm based on Received Signal Strength Indicator (RSSI) is easy to introduce singular points when locating indoors, which leads to the decrease of positioning accuracy. To solve this problem, this paper proposes a Similarity-based K-Nearest Neighborhood Location Algorithm (SKNN). Referring to the idea used to solve the problem of similarity of nodes in bipartite networks, this algorithm builds a bipartite network between the location fingerprint and the AP. It proposes a similarity parameter which can be used to modify the K-Nearest Neighborhood localization algorithm. The experimental results show that the SKNN algorithm proposed in this paper can effectively reduce the influence of singular points on the positioning results and improve the positioning accuracy, with 80% of the positioning errors within 2m, and the effect is obvious in the large scene.

Abstract:The direction of arrival estimation has been widely employed in Wireless Sensor Networks. This paper proposes a Weighted Subspace Fitting algorithm which can largely reduce the computation amount when doing high-dimensional non-linear optimization by limiting the genetic searching space. This method uses rotation invariant subspace and unbiased estimator of the theoretical minimum error to limit the search space and the complexity of WSF algorithm is reduced by shortening the genetic length of genetic algorithm. The simulation results show that this algorithm has the same performance as the WSF algorithm. Compared with other intelligent optimization algorithms, the proposed algorithm significantly reduces the computational complexity of the algorithm.

Abstract:With the widespread use of personal mobile communication devices and location-aware devices, the mobile communication service provider has accumulated a lot of its users' location data. At present, most researches on location data are focused on the mining of active trajectories. A small amount of researches on the pattern of personal stay only determine activity stops, but lack further mining. We conduct researches based on the base station data and propose a simple method to identify the activity stops according to the characteristics of the base station data. Then we propose two methods for mining the pattern of personal stay. Finally, the real data are used to verify the effectiveness of the algorithm.

Abstract:Aiming at the problem of detecting the centers of touching particles images, a detecting method based on the improved Generalized Hough transform is proposed in this paper. The method firstly figures out the ring that overlaps the known particle's outline, then the overlapping area of the ring template is accumulated along the foreground contour of the image to be detected. Finally, the region maximum value of the accumulated matrix will be the centers of the particles. The ring template has rotation invariance so that the detection time will be shortened greatly. Meanwhile, the result shows that the method in this paper has fine detection effect.

Abstract:In view of the characteristics of traffic networks capacity under the situation of mass emergency, this paper cites the BPR impedance function to solve the vehicle travel time between the various sections. It builds the shortest path and the shortest vehicle scheduling model and designs the improved discrete Glowworm Swarm Optimization Algorithm. It constructs a numerical example to solve the model and the results are compared with the results of genetic algorithm. The feasibility of the algorithm is verified and can better meet the needs of emergency rescue vehicle scheduling.

Abstract:A heuristic swarm intelligent optimization algorithm for coalitional resources games in virtual enterprises is proposed by comparing the similarity of coalitional resources games in virtual enterprises with the classic SAT problem. The algorithm fuses portions of the principles of Glowworm swarm optimization algorithm and Cuckoo Search optimization algorithm. It designs the feasible cross operator and the mutational operator, which can repair infeasible solution and maintain diversity. Experimental results indicate that the algorithm's iterations linearly increase with the stable coalitions ever found. Compared with the heuristic genetic algorithm, it performs better in hill-climbing performance and searching efficiency.

Abstract:The traditional classification system often chooses the Naive Bayesian algorithm as the classification algorithm. In the course of the study, we find that the Naive Bayesian model(NBC) has the following conditions:all the characteristics do not mutually influence each other, and the feature attribute weights is 1. But we find that is not the case after a study. In the classification of data, some features may have a greater impact on the classification results, while some may have little impact. In order to optimize the algorithm, we need to attach different weights to different features, so as to obtain the classification results more objectively. This paper studies two kinds of calculation methods of attributing weight based on the traditional algorithm. At the same time, considering the characteristics of mobile phone forensic data, it proposes the calculation method of two kinds of improved weight suitable for mobile phone forensic data. This paper researches the improvement principle of research, compares the improved calculation method of weight with the traditional calculation method in their impacts on the classification results using the same classification algorithm with the same data.

Abstract:The ray casting is a widely used basic volume rendering algorithm. It can get high quality image but the computation of traditional sampling-points method is overlarge and the rendering speed is very slow. Aimed at the shortcoming, an algorithm that takes the advantage of the recurrence relation of parallel projection line during the sampling process is presented to improve the speed of obtaining the sampling-points and the speed of reconstruction. Finally, the algorithm is realized on PC platform. The result shows that compared with standard ray casting algorithm, the accelerate algorithm does not only improve the rendering speed for nearly ten times but it does not reduce any image quality. It gives a more capable method for the application of medical image 3D reconstruction.

Abstract:In this paper, we adopt the Gabor transform to extract features of facial expression images and use a series of locally linear embedding(LLE) algorithms to reduce the data dimension. The LLE algorithm widely used in facial expression recognition is a kind of nonlinear dimension reduction algorithm. It is able to make dimension-reduced data keep the original topology. Because the LLE algorithm does not take the category information of samples into account, the supervised locally linear embedding(SLLE) algorithm appears. But the SLLE algorithm only considers the category information of samples, and does not take the relationship among various expressions into account. Therefore, in this paper, we propose an improved SLLE algorithm, which regards the neutral expression as the center of the other expressions. The results of facial expression recognition experiments on the JAFFE database show that our algorithm obtained better facial expression recognition rate compared with the LLE algorithm and the SLLE algorithm. Our algorithm is more effective

Abstract:In recent years, the convolution neural networks in the field of pedestrian detection have achieved similar and even better results, compared to other methods. However, the slow detection speed can't meet the realistic demand. To solve this problem, a real-time pedestrian detection method is put forward. The scattered detection processes are integrated into a single depth network model. Images which can be calculated through the model can directly output detection results. The extended ETH dataset is used for training and testing the model. The experimental results show that the method is very fast and can achieve the goal of real-time detection with the guaranteed accuracy.

Abstract:This paper proposes a novel image inpainting algorithm based on exemplar repaired in a sequential manner. This method makes a new experiment on inpainting order of Criminisi classical algorithm. The inpainting order of the classical algorithm is obtained by calculating the priority. But the priority is tending to 0 with the sustained inpainting. As a result, the algorithm will fail and produce error inpainting results. In order to solve this problem, we replace the priority determined order with a sequential manner, which can keep the algorithm working all the time. Moreover, we propose a novel exemplar pattern like an inverted L. The pattern in our method can help to enhance the ability of structure propagation and improve correction rate of patch matching. Experimental results prove that the novel algorithm can achieve better visual effects.

Abstract:Web service could make events and data generated in software products as service by which software products interact with each other. Workflow is a very popular way to deal with message flow and event flow which could deliver events and data among software products. It is very meaningful to study Gitlab which is known as a widely used open source code and document management tool. However, poor flexibility, weak expandability and raw service granularity are found in Gitlab service during the study. To tackle the problems in Gitlab, this paper proposes a new service solution to Gitlab which redefines the service process, service standard and service implement. To implement the services, we design and implement new solutions for listening services and execution services which involve the message queue and asynchronous mechanism. According to experimental analysis to this solution, GITService has high flexibility, strong expandability and intensive service granularity with little cost of time. The solution we provide in this paper is of significance to design and implementation of service in other situations.

Abstract:Abnormal behavior detection has a wide application prospect in the field of self-service banking intelligent monitoring system. In this paper, an anomaly detection method based on regional optical flow feature is proposed. Firstly, the mixed Gaussian model is used to represent the change of the background pixels and the background model is updated. The motion foreground is extracted from the video sequence with the background difference method. The optical flow information in the moving region is calculated with the lucas-kanade optical flow method. The weight-oriented histogram is used to describe the behavior, and the motion anomaly region of the histogram is calculated by using the motion entropy of the histogram. Then the SVM is used to classify the anomaly regions. From the experimental results, it can be seen that the abnormal events can be identified better and the real-time performance is better, which can meet the practical application requirements.

Abstract:Software Defined Network (SDN) provides a new solution for network virtualization, which can virtualize a set of infrastructures to multiple logical networks to meet different network requirements. This paper studies the mapping of virtual network rules to physical network rules when multiple physical switches are virtualized into one big switch in SDN network virtualization. Considering link load, rule distribution and node load, this paper proposes a three-stage rule mapping optimization algorithm. First, according to the rules of the virtual network, the multicast source node and a destination node set are located, and a rule mapping tree is generated by using the MPH algorithm. Then, the rules of the virtual network rules are deployed to multiple nodes of the physical network. Finally, with the consideration of the node load, the rule deployment is finely-tuned to generate the virtual rule mapping strategy. Simulation results show that compared with direct egress node deployment the average number of network node rules is reduced by more than 40%.

Abstract:On an OpenStack cloud platform, a server may simultaneously run more than a dozen virtual machines, which requires high system network I/O performance. Therefore, the efficiency of I/O virtualization is important for the improvement of network performance on OpenStack cloud platform. In order to improve the overall network performance of the system, introducing SR-IOV technology into the OpenStack cloud platform is an option. The influence of SR-IOV technology on network I/O performance of OpenStack cloud platform is tested by contrast experiments. Finally, the experiment results show that, after the introduction of SR-IOV technology, the I/O virtualization performance of computing nodes in OpenStack cloud platform have increased by about 50%.

Abstract:This paper discusses the process of applying code audit to analyze the vulnerabilities of OpenSSL source codes and proposes some specific fixing advice for OpenSSL. Source level analysis mainly contains data flow analysis, dynamic taint analysis and path constraint solving proof method, etc. Because various code audit techniques adopt formal analysis on software architecture based on their own security requirements, they usually produce good effects when aiming at some particular scenes, but they lack universality. When auditing important mature projects like linux and xen, it is even impossible to exploit vulnerabilities efficiently with using these code audit techniques with high false rate. In this case, the collocation use of different code audit techniques is applied, as well as a new method of the security attributes definition from the bottom to improve the accuracy of software security requirements description and to avoid the defects in its audit. These methods increase audit efficiency, decrease false positive and process deep vulnerability exploitation while retaining the advantages of the high degree of automation of code audit.

Abstract:At present, the application of computer virtual technology has become one of the important trends in the field of technology. This paper analyzes the specific needs of virtual machine technology to discuss three key points-the simulation of measuring machine, measuring path and virtual measurement data from the construction of virtual machine technology, including difficulties. It provides the specific ideas and methods. Finally, according to the function division of the virtual measuring machine module, it builds a virtual measuring machine prototype, and verifies the feasibility, ideas and methods, which are effective and available on the HOOPS platform.

Abstract:Aiming at the problems that the machine sketch image drawn by the manipulator lacks realism, artistic sense, and the hardware equipment is complicated, and hard to popularize, this paper proposes a machine sketch method based on image hierarchy. With this method, it extracts image contours and image inner filling areas as binary target images from the processed images based on the application of some pre-processing methods such as Gray-scale Transformation, Filtering and Binary in dealing with the original character image. Through traversing the target images with simple path planning, then using XY-type simple manipulator to draw the image contours while painting the images with hierarchical drawing modes, it ultimately achieves the drawing of the character sketch image. Experimental results show that the method can draw arbitrary curve according to the image contours, through the hierarchical drawing model, which take the environmental factors into consideration, achieving the purpose of flattering the sketch image and making it more hierarchal.

Abstract:With its rapid development, WeChat has become an important platform for interpersonal communication, and occupies a unique position in the social network information dissemination. Based on the epidemic model, WeChat information propagation has established. The influence of model parameters and information sources on the scope and characteristics of WeChat information propagation is analyzed in multi-scenario with experimental data. Model simulation results and characteristics of propagation of WeChat have a very anastomotic match, which shows the WeChat information dissemination effect is important to control the use of WeChat.

Abstract:With the development of software internationalization technology, internationalization testing research is attracting more attention. In order to solve the problems of hardcoding and over-translation in international software, this paper proposes an automatic testing scheme. By analyzing the concept, classification and testing methods of the two problems, this paper summarizes the shortcomings of existing manual testing and the characteristics of Struts2 internationalization technology. Finally, combined with the actual situation of the tested project, the automatic test scheme is elaborated. The scheme has been applied to the internationalization testing of a real Web product, which is proved to be the support to hardcoding and over-translation automatic test, and has achieved good results.

Abstract:In view of the low efficiency and high labor intensity in agricultural greenhouse, in order to improve the automation of agricultural greenhouse, an automatic spray robot based on STM32 MCU is proposed in this paper as the core controller. The robot detects lines based on gray sensors. A three-wheel differential rotation is used to control its traveling and turning. A planning algorithm suitable for greenhouse environment of agriculture is presented based on grid method. Pump switches in water tank are controlled indirectly by an electric relay, and spray process can be controlled via bluetooth remote control. Temperature and humidity sensors are used to feed real-time measurement and moisture of soil. Meanwhile, these messages are sent back for processing. Experiments show that the spray robot which is designed based on modularization is fully functioned to promote the use of agricultural robots.

Abstract:Based on the intelligence of the finance experts and the knowledge engineering researchers, this paper tries to analyze and summarize the internal rules of the big data of market information, and to construct the finance ontology base and the ontology reasoning rule repository of the trends of economic trends, through the various market factors of the economics of China. It uses ontology reasoning technology to translate the experience from the finance experts and the objective rules of economic development into machine readable and computable ontology reasoning rules, based on which ontology reasoner performs economic trends prediction.