Abstract:According to the development of support vector machine, this study reviews many literatures based on applications in different domains, such as text classification, human body detection, vehicle traffic recognition, medical examination, and so on. Meanwhile, the theory and development of support vector machine are both expounded in detail from the principle of kernel function and its multiple classifications based on actual dataset. The potential improvements of support vector machine technology are infinite. We look forward to see their development prospects.

Abstract:It is of great significance how to model and mine historical data quickly and effectively. Based on the statistical characteristics of Markov model, this study designs and implements a variable order Markov model based on suffix array and suffix automata, in view of the limitations of the model in practical data mining applications. Based on the realization of suffix tree structure, the suffix chain is introduced to realize the quick jump of each state subsequence, and the requirement of different order length probability can be dynamically and adaptively calculated. The experimental results show that compared with the traditional Markov model, the model constructs the link between suffix sequence characteristics of probability and statistics of historical data and the state in linear time and space complexity, which can greatly reduce the storage space and time, and realize online learning and application of large data.

Abstract:The self-organizing network has the capability of actively adapting to the change of space environment. Therefore, it has a wide application prospect in space communication. This study analyzes the especial space resource for space self-organizing network, and according to the distance among space objects, the density and application scenarios space self-organizing networks are classified into four types including spatial-terrestrial self-organizing network, intra-satellite self-organizing network, inter-satellite self-organizing network, and interplanetary self-organizing network, whose networks performance are compared and analyzed. Finally, the directions of research in space self-organizing networks are introduced including network structure, network protocol, security management, and router protocol. A space network structure is suggested for local automatic and whole collaboration based on the mechanism "preparation-storage-forward- handle".

Abstract:In the background of big data, the relationship presented in data is not intuitive, so the visualization of the relationship is of great practical value for extracting implicit knowledge. Considering the edge-node relationship of the visualization of character relations and the visualization requirements of displaying the relevant information, the D3.js was improved to optimize the links between edges and nodes, which is convenient for data processing, especially for dynamic data. The shortest path algorithm was improved to solve all the minimum rings containing the specified nodes of undirected graphs, and applied to seeking groups in the relationship map. The data-driven features of D3.js visualization library were taken advantage of to display the character relationship map with data exchange function.

Abstract:In e-commerce, as one of the most secure protocols, the SET protocol resolves some security issues. However, SET encryption and decryption program have gradually been doubted because of its lack of security guarantee. An improved elliptic curve cryptography is proposed to replace the original private key encryption algorithm which can improve the speed, performance, and security of the protocol. In this study, an improved NAF algorithm is proposed to develop the encryption and decryption speed of data affected by time-consuming ECC point multiplication. Compared with existing algorithms, the improved algorithm has better time complexity and it uses less computational resources. Besides, the MD5 hash generation algorithm is used to further improve the security of the existing elliptic curve cryptosystem.

Abstract:The rapid development of Internet has promoted the extensive application of e-commerce. The stolen identity of users, and the electronic contract tampering have seriously affected the fairness and security of online electronic transactions. For the main participant of the transaction, it is a primary and pivotal problem to solve how to confirm the identity of the other party is genuine and credible and how to confirm the electronic contract is sent by the other party with authenticity, credibility, and non-repudiation. In order to ensure the uniqueness of the user identity, transmission data integrity and non-tampering, the contract process traceability, this study combines the user authentication, electronic contract encryption transmission, notary office to participate in notarization and other methods to design electronic contract service platform, which achieves fairness in the use of the platform.

Abstract:In order to ensure the daily examination and maintenance of the river water ecological environment more efficient, the intelligent river management system based on mobile terminal has been designed and implemented. An efficient track rectifying algorithm is adopted in this system, which makes the track more smoothly displayed in the map by filtering out the mistakes and invalid track positioning data. The index scale-based analytic hierarchy process in rank model of river pollution is proposed, and the rank of river pollution in the region is sent to the user by push technology. The results demonstrate that the system operates with simplicity and stability, and improves the efficiency of river staff work, solving the problems that the staff can not give timely treatment to the inspection tasks problem as well as whether the staffs follow the scheduled timetable and route.

Abstract:In the emerging growing of bilateral exchanges between China and Thailand and the widespread use of Android apps, the study has designed and implemented Chinese-Thai-English translation audible electronic dictionary APP on the Android platform. The application is designed based on Android Studio development environment and it uses Java language and SQLite database. This study creates a local word library from the Thai language corpus in a special way, and solves the problem that SQLite visualization tool processing Thai will appear garbled. The key technology of the system is to use SQL language to inquire word meaning in the created local word library. The system has the features of dialogue translation and photo translation, and it also has the functions of trilingual query translation among Chinese, Thai, and English, and the pronunciation of Thai language by real people. The test shows that the software has certain convenience and practicability.

Abstract:In recent years, a large number of video surveillance systems are deployed in the nature reserves, so it has become an urgent problem how to effectively use the increasing mass of video surveillance data. In this study, an efficient algorithm for key frame extraction based on image similarity is used to clean and compress the massive video data. At the same time, an object detection algorithm based on deep learning is used to extract valid video information. In addition, the system provides a variety of content-based video retrieval methods. It automatically analyzes and processes the search contents submitted by the user so as to quickly retrieve the video of interest. This study analyzes and retrieves the video surveillance data of wild animals in Qinghai Lake, which verifies the correctness of the proposed system.

Abstract:Firstly, this study expounds the advantages of frontend-backend separation development of web application, introduces the MVVM design model, and points out the necessity of designing dynamic forms. Then, it describes the specific design of dynamic forms from the frontend to the backend in details. The development of the frontend is based on the lightweight framework of Vue.js taking advantage of the features of its components. The backend uses Spring MVC architecture to implement the business model and data process and utilizes NoSQL database to satisfy the dynamic requirements of form items and fields. At last, this study describes a responsive design about terminals self-adaption.

Abstract:From the perspective of virtual simulation teaching, this study takes the modern farm as the object of reference and designs the simulative teaching system of "Agricultural Mechanics" based on the Unity3D engine in order to improve the situation that it is difficult for teachers to teach and for students to learn the agricultural machinery in a traditional way. This study demonstrates some of the original developing thoughts, function and structure of the system, and introduces the key technologies involved in the development process. The application result shows that the system realizes the virtual visualization of agricultural machinery equipment in the process of production operation in virtual farms, and displays online the virtual dismounting and assembly, maintenance and operating principle of all kinds of farm equipment. It has certain practicability and teaching value.

Abstract:This study takes the background of improving the quality of video meeting for an enterprise, guided by the principle of advancement, convenience, and flexibility. Using high definition video conference terminal equipment and MCU, the system is designed to be a complete HD video conference system across all levels of units. To enable the double video stream to have mainstream HD display capabilities and to be compatible with other resolution modes, it meets the needs of video conference communications. Finally, the results show that the system is feasible and reliable.

Abstract:In the research of modern urban planning, the in-depth analysis of the information that focuses on human is crucial. The use of an effective video analysis technology to analyze and monitor video can greatly expand the basic data of pedestrian, which is of great significant to urban quantitative researches. This study deals with video that shot pedestrians in the same street for a period of time. Deep learning is used for detecting pedestrians in the specified monitoring area of the video based on the forward propagation convolution neural network model. In order to ensure the accuracy of the information for pedestrians, it tracks the detected pedestrians and determines whether the target is lost. Finally, it quantifies the number of pedestrians, the direction and speed of movement, the time of retention, etc., and carries out corresponding data analysis. The results show that the method can effectively quantify data of pedestrian information, then provide accurate and effective data support for urban quantitative studies.

Abstract:With the development of computer technology, the intelligent and automatic pedestrian detecting system continues to emerge. The pedestrian detecting system is an important part of the information management for enterprises and institutions. The traditional method has a lower accuracy due to limb occlusion and light changes. This study proposes a vertical detection method for the head features, which can ensure that the features cannot be blocked even in high flow density. Firstly, the gradient histogram feature of the foreground image is extracted and the head target is detected by SVM. The MeanShift algorithm is used to track the head target in the adjacent frame by using the color feature of the head. The pedestrian is detected according to the target track. The algorithm is applied and tested in the embedded system. The experiment results show that the method is effective and accurate.

Abstract:Aiming at the problems of the traditional desktop, a virtual desktop solution based on the SPICE framework and Openstack platform is proposed. The server cluster is built with KVM virtualization technology, and the SPICE protocol connects the client and the server to transmit data. This study analyzes the current mainstream transmission protocol and the key factors that affect the performance of the desktop, and introduces the realization process of the virtual desktop in detail. The test proves that the virtual desktop has good user experience and practicality, and it effectively solves the problems of the traditional desktop in management and cost.

Abstract:In view of the limitations of the traditional fundus camera with poor mobility, complicated equipment and image transmission limited by wiring and object pose, a wireless retinal imaging system based on WiFi wireless local area network is designed. The system mainly consists of three components, including the imaging system, PC software operating system, and image transmission system. A hand-held fundus camera realizes real time images acquisition, and displays the image information on the PC side through wireless transmission mode, to achieve real-time observation and preservation of the fundus image and as an objective basis for the diagnosis and treatment of ophthalmic diseases. Finally, the examination results will be stored in the database to establish electronic medical records with relational database SQL Server 2008, so that the results can achieve the functions like archiving, accessing, and sharing for the diagnosis of retinal disease in remote medical and large data mining applications.

Abstract:To simplify the computational complexity of intra prediction mode selection for H.264/AVC, a fast intra prediction mode selection combination algorithm based on macroblock judgment is proposed in this study. This algorithm uses the three points gradient operator with a low-complexity to improve the edge direction histogram intra prediction mode selection algorithm (Pan algorithm), which obtains edge orientation vector by Sobel operator. Firstly, the MAD (Mean Absolute Difference) value is used to predict the threshold of the encoded macroblock type combined with QP. Then, the improved Pan algorithm is used to filter the pre-determined macroblock. Finally, an optimal prediction mode can be determined. The experimental results show that under the condition of six video sequences with all-I-frame coding, this algorithm reduces the coding time by about 72.4%. Compared with the Pan algorithm, the coding time is reduced by 28.6%, while the bit rate is increased by 4.21% and 1.8% respectively, and the PSNR is invariable.

Abstract:A probabilistic coverage decision-theoretic rough set (PCDTRS) model is proposed in this study to deal with the two main issues in rule acquisition from decision table, i.e., contradiction of extracted rules and redundancy of override sample. Firstly, the basic theories of the decision-theoretic rough set (DTRS) model including the attribute and value reduction algorithms are presented. Subsequently, the probabilistic coverage model is raised based on the DTRS model, and three levels covered matrixes meeting the needs of value reduction are proposed to resolve the aforementioned problems. Finally, the results of a series of experiments on Chinese cookbook nutrition illustrate the feasibility and effectiveness of the PCDTRS model. Compared with other models, the reduction strength and the number of conflicting rules using the PCDTRS model are higher and fewer respectively.

Abstract:Taking the non-equilibrium distribution characteristics of the coal mine water burst sample set into account, this study presents a coal mine water inrush prediction model based on the integrated learning classification. It focuses on the construction method of base classifier, the performance index and the weight analysis of base classifier, and the integrated learning algorithm based on improved Boosting. The experimental results show that although the algorithm does not achieve the minimum error rate of non-waterlogging samples, a 100% discrimination rate for water burst samples is realized, and the calculation load is small and it is easy to realize.

Abstract:To improve the performance of Chinese text classification, a rule matching method based on rough set theory is proposed in this study. In the extracting process of textual features, the CHI statistical method is improved and the weight of the feature is scaled and discretized. It combines the discriminant matrix to achieve the attribute reduction and rule extraction for rough set theory, and uses rule pre-test method to optimize the decision parameters of rule matching to improve the effect of Chinese text categorization. The experimental results demonstrate that the categorization accuracy of the improved matching method is higher, and in the case of less training data, it can also achieve decent results

Abstract:With the continuous growth of Internet users, the high concurrency becomes a major challenge in building large-scale electricity-business website system. To solve the problem, the load-balancing algorithm is used to realize the balanced load of each node in the Web service cluster. However, the current load-balancing algorithms generally have some shortcomings. In view of this problem, this paper proposes a dynamic adaptive weight round-robin random load-balancing algorithm (DAWRRRLB). This algorithm takes into account the multiple factors that affect the performance of the server nodes in the Web service cluster, and changes the load performance of nodes in the cluster according to the node in the operation process of the real-time dynamic load. It combines with the Pick-K algorithm to improve the dynamic adaptive weight round-robin random load-balancing algorithm, ensuring the best performance of the server node is providing services. By many experiments, the DAWRRRLB algorithm is proved to be able to effectively improve the load-balancing efficiency.

Abstract:In order to select a suitable asymmetric polling service in the wireless sensor network, this paper analyzes the performance of the asymmetric gated service and the exhaustive service and compares the superiorities of the two services in different situations. Generally, in analysis of asymmetric polling services, a progressive analytical method is usually utilized. Therefore, the service model of the two queues will serve as a basis, and then expands on this basis to analyze the asymmetric service of multi queue. The mathematical model of the service system is constructed by using the Markov chain and the probabilistic parent function in the analysis process. Through the analysis of the mathematical model, the expressions of average queue length and average query period of asymmetric service system are given. According to the comparison between the theoretical value and the experimental results, it can be verified that the two are consistent. In addition, in the wireless sensor network to achieve the asymmetric gated service and the exhaustive service for the initial design, it can achieve multi-hop routing protocol, into a single-hop polling protocol to reduce the conflicts of data transmission.

Abstract:In view of the problems that when processing massive data the traditional K-means is highly complex and insufficient in computation, a SKDk-means (Spark based kd-tree K-means) parallel clustering algorithm has been proposed. The algorithm improves the choice of initial center point by introducing kd-tree and overcomes the problem that the traditional K-means algorithm is easy to fall into the local optimal solution due to the uncertainty of the initial point. During K-means iterative calculation, the redundant computation has been reduced and clustering speed has been accelerated by the nearest neighbor search of kd-tree. The parallelization of the algorithm is realized on the spark platform and it is applied to the massive data clustering. Finally, the experimental results show that the algorithm has good accuracy and parallel computing performance.

Abstract:With the development of mobile networks and intelligent terminals, the recommendation of the companion based on high-quality users has become one of the hot topics in the Internet, and the recommendation algorithm about companion is the crucial factor. In the past, the user location trajectory similarity recommendation algorithm was mainly based on geographic location or base station data and the data sparse may result in undesirable results. This paper proposes a companion recommendation model based on the cosine similarity of IP sites. More comprehensive IP sites data have been used instead of geographic data, and the date time data are calculated for cosine similarity to eliminate the data sparseness problem. Finally, the people with higher similarity and higher quality are recommended.

Abstract:Thermal comfort is an evaluation index of indoor environment comfort. Since the calculation of thermal comfort is a complex nonlinear iterative process, it is inconvenient to apply to air conditioning real-time control system. In order to solve this problem, use the BP neural network algorithm to predict thermal comfort. However, in order to improve the slow convergence rate of traditional BP neural network, the bird swarm algorithm (BSA) is used to optimize the initial weights and thresholds of BP neural network. Finally, the BSA algorithm is compared with the similar particle swarm optimization (PSO) algorithm. MATLAB software is used to simulate, and the simulation results of BSA-BP prediction model are compared with the simulation results of the basic BP neural network prediction model and the PSO-BP prediction model. The results show that the BSA-BP algorithm has faster convergence speed and higher prediction accuracy.

Abstract:Defects like bubble and solid inclusion may disrupt the photovoltaic manufacturing process. The reliability to detect the defect of Photovoltaic Glass is particularly important. In order to separate defects from periodic background, this study introduces a novel approach to estimate saliency using image contrast and image signature. On the one hand, candidate saliency is computed based on center-surround contrast. On the other hand, original image is firstly converted by DCT transform, then the sign operation is applied to produce the image signature and the reconstructed image is computed by IDCT transform. Later the Gaussian kernel is applied to gain the reconstructed saliency. Finally, the reconstructed saliency is used to fuse the candidate saliency. Experiment results show that the proposed method gains relatively accurate saliency regions compared to the 7 methods.

Abstract:The stochastic strategy exists in LPA, which seriously destroys the robustness of the algorithm. With the advent of big data age, the scale of complex networks is increasing, which causes the computation of the algorithm to increase and the convergence rate to slow down. A new improved label propagation algorithm-KLPA is proposed to solve this problem. Firstly, the network is preprocessed by using the K-Shell index to divide the network into a core-edge layer, remove the nodes of the edge layer, and assign labels to the nodes in the core layer. Secondly, the improved propagation strategy is used to divide the community for preprocessing network. Finally, experiments show that the KLPA algorithm reduces the size of the network, effectively improves the quality of community division, and accelerates the convergence rate of the algorithm.

Abstract:The global image detection of feature points is time-consuming, and the global feature is not of good of stability, which causes the algorithm speed to be slow and the matching accuracy to be low, with the matching effect satisfactory. On the basis of scale invariant feature transform (SIFT) based on the sparse structure of the concept, this study puts forward an image feature matching algorithm based on sparse structure (SSM). It gets the pixel value by sparse sparse degree function, selects pixel highly sparse region, and detects the SIFT feature point of the region, to achieve feature matching by using the best descriptors. Compared with several classical algorithms, the experimental results show that this algorithm has significantly improved in feature matching speed and accuracy, and it can be used for real-time object tracking, image retrieval and image mosaics, and other fields.

Abstract:Data fusion is a method to improve the calculation efficiency and reduce redundant data. The air temperature data of Xinlinhe basin is carefully researched. Aiming at the drawback of traditional Kalman filter approach:a slight fluctuation, a novel method is proposed based on the traditional Kalman filter and distribution map to fuse the air temperature data. The task is to make the data collected every five seconds fuse into the air temperature value of an hour. For the demonstration the proposed method, disturbance data and mutation data are set on the basis of the original data. Via the experimental simulation, the improved algorithm has a good fusion effect, with strong anti-interference and stability, which can raise the accuracy of the meteorological data.

Abstract:Focusing on the issue that the correlation filter tracking algorithm under the condition of long-term occlusion or scale change has poor performance, the proposed algorithm makes the improvement based on the kernelized correlation filters tracking method. Firstly, the histogram of gradient and color-naming of target area are fused to construct training samples in order to improve the description of the target. Then, the scale is obtained by calculating the maximum response on the multi-scale image pyramid. Finally, the re-detection mechanism is introduced, and only when the response of the target is less than the threshold, the online random fern classifier is trained to re-detect objects. The obtained results of experiment demonstrate that the proposed algorithm is robust in the tracking of fast motion, heavy occlusion, out of view, and other complex scenes.

Abstract:The gravitational search algorithm (GSA) is a relatively novel swarm intelligence optimization technique which has been shown to be competitive to other population-based intelligence optimization algorithms. However, there is still an insufficiency that is the low convergence speed of the standard gravitational search algorithm, and its being stalled easily in the evolutionary process. Considering those problems, an improved gravitational search algorithm is presented. A strategy of chaotic opposition-based learning is employed to generate an initial population, which makes it possible for the algorithm to achieve a better initial population, thus accelerating the convergence speed. In addition, the method makes full use of the exploration ability of the search strategy of artificial bee colony algorithm to guide the algorithm to jump out of the likely local optima. The results of numerical simulation experiment on a suite of 13 benchmark functions demonstrate the effectiveness and superiority of the improved gravitational search algorithm.

Abstract:In view of the traditional need to calculate the curvature and the insufficient threshold selection of detection algorithm angular profile curve, this study proposes an angle detection analytic algorithm based on Freeman code by first edge detection, contour extraction to give Freeman code profile. When the chain code changes, it analyzes whether the chain code of a number of points before and after it is consistent with certain rules to determine the corner point, without the need of traditional corner threshold selection, curvature calculation and other steps. Compared with He & Yung, CPDA, Fast-CPDA, and ARCSS corner detector, the experimental results show that the proposed algorithm has the highest accuracy (ACU) in detecting the corner. The transformation experiments show that the average repetition (AR) is the highest, which shows that the proposed algorithm has good corner detection performance.

Abstract:The images combining visible and infrared light would produce color deviation. In response to this phenomenon, this study proposes a de-crosstalk matrix color correction algorithm based on polynomial regression to counter the crosstalk problem of four-band images (RGB three-band and IR near-infrared band). Under specific lighting conditions, the algorithm uses 24 color blocks of standard color checker to build the de-crosstalk matrix according to the idea of polynomial regression. The de-crosstalk matrix will be used for four-band images which are taken under the same lighting conditions to remove the near-infrared crosstalk and realize color correction. The experimental results indicate that the method does not only need to calibrate the correction matrix at one time, the color deviation images taken under the same lighting conditions will be corrected and restored the natural colors of themselves thereafter.

Abstract:Because of the neglect or misuse of the attribute weight of the existing intuitionistic fuzzy clustering algorithm, an intuitionistic fuzzy clustering algorithm based on the intuitionistic fuzzy analytic area is proposed, which introduces the analytic area of intuitionistic fuzzy sets and a method for calculating the attribute weights, then constructs the objective function of clustering and gives the clustering procedure. Finally, an example shows the validity and feasibility of this algorithm.

Abstract:With the development of computer technology, the processing capability of CPU with traditional architecture cannot address the various computing tasks. There is still room for the performances of heterogeneous computing to be promoted. This study analyzes the idea of "applications make structures, structures make efficiencies", clarifies the conception of mimicry computing, which is based on multidimensional reconfiguration function and dynamic changeable operation system, and then uses the characteristics of programmable hardware, dynamic reconfiguration, and low power of FPGA to design a mimicry computing server which is based on FPGA, and also utilizes the core circuit and critical technology of this mimicry computing server.

Abstract:The information dissemination can be affected by social reinforcement effect. Users usually can not receive the information in a short time, and when they do not receive information for a long time, the information dissemination rate will decrease because of the forgetting mechanism. Therefore, it is believed that when information dissemination is affected during the social reinforcement effect, it is also affected by the forgetting mechanism. Information dissemination rates with impact of social reinforcement effect are set as initial values of forgetting mechanism in this study. SEIR model is improved and the transfer of node state in SEIR model is redefined. Simulation results indicate that the larger of time interval for receiving information, the lower of the rate of information dissemination, and the range of information dissemination will be smaller when taking the forgetting mechanism into consideration during social reinforcement effect in social networks.

Abstract:The improved particle swarm optimization (IPSO) optimized hyper-sphere support vector machine (HSSVM) can be used for abnormal detection of keyboard in this paper. Firstly, the development of the hook (hook) procedure in the Windows operating system is used to collect the required key time series as a training set and test set through the system messages WM_KEYDOWN and WM_KEYUP capture keyboard keystroke messages. Then, the HSSVM model is used to carry out sample training and finally transformed into a quadratic programming problem. The IPSO is used to optimize the penalty factor and kernel parameters of HSSVM model. Finally, the test set is used to verify the accuracy of the model detection and is compared with the results before optimization. The test results show that the IPSO-HSSVM model is effective for the detection of the keyboard and the accuracy rate is over 90%, which is better than that of the HSSVM before optimization. However, it is necessary to further improve the quality and quantity of the training samples in order to obtain higher detection accuracy.

Abstract:Aiming at the practical needs of multi-user collaborative editing in power grid GIS, a multi-user cooperative editing of power grid GIS based on locking mechanism is proposed based on the task division and version management technology to ensure the consistency and correctness of the edited data. This study first analyzes the architecture and editing process of collaborative editing. Then it focuses on the implementation of the locking mechanism in the collaborative editing, the locking process and the data synchronization between the clients. Finally, a prototype system is constructed based on the test data of a certain province, and it verifies the multi-user collaborative editing of power grid GIS based on locking mechanism.

Abstract:With the increasing demand for human-computer interaction, the vision-based gesture recognition has attracted much attention in many fields. The depth image is more and more popular in target recognition field for its good performance. The gesture region is divided from the depth image and normalized to obtain a unified specification gesture binary image, and then it detects the gesture edge. A progressive Hough transform algorithm is proposed to detect the finger edge curve and extract the finger information in the gesture image. Then it extracts the features based on the edge curve and accords this to establish 3D histogram. Finally, the two features are fused, and the gesture classification is carried out by the minimum closed ball support vector machine (MEB-SVM) according to the obtained eigenvector. And the recognition rate on the test set is 96.6%. The new method does not depend on color, detail texture, and other information, with good robustness. And the method is faster to meet the needs of general applications.

Abstract:The migraine is a common disease with high incidence. It is still not easy to explain its pathogenesis very well. Therefore, it lacks effective diagnostic methods. This study aims to predict migraine by using the functional magnetic resonance imaging technology to obtain functional network of brain, then through deep learning of automatically it extracts data features by Autoencoder, combined with various machine learning algorithms to provide a reference for clinical diagnosis of physicians. It can get better classification effect to extract data features and train the classifier by deep learning. The deep learning algorithm, based on the initial features obtained by the traditional templates, can further extract more fine and effective features, and obtain better classification performance in predicting migraine.

Abstract:Since the introduction of software engineering concept in 1968, it has experienced the development of nearly 50 years. The scale and complexity of software system have been increasing day by day. However, since the 1970s, there has been a delay in the progress of a large number of software projects. And quality defects for the typical characteristics of the software crisis, is still frequent. Software cost estimation in the software engineering development process has been playing an important role. The precise software cost estimate is the guarantee of software engineering on time. In this study, a similarity measure based on Pearson correlation coefficient is used, and the software cost estimation is carried out by using the TOPSIS method to estimate the cost of the project with the closest project. Finally, the method is applied to the Desharnais dataset and compared with other methods. The experimental results show that the software cost measurement method based on the correlation coefficient has better accuracy than the existing similarity measure method.

Abstract:Considering the outdoor monitoring system using the image to identify weather conditions, a novel classification method with bag of words model is proposed which combines SVM with random forest to identify the sunny or cloudy days. The bag of words model uses the SIFT feature to construct the dictionary by clustering, and uses the least squares method to solve the dictionary structure parameters of the optimal image. Finally, the multi-scale image bag of words model feature is obtained according to the pyramid matching. The construction of the classifier uses the support vector machine (SVM) as the first level classifier to classify the small confidence samples, and then uses the random forest as the second level classifier to judge. Through the test of the 10 000 images of two categories of weather images, the recognition accuracy verifies the effectiveness of the method.

Abstract:Aiming at the interaction in the actual prototype system, this study analyzes the relationship between the level and the other layers. It analyzes and complements the delivery path of the hardware instruction, and proposes an implementation method based on the improvement and expansion of the open source game engine cocos2d-x, which is stored in bits by capturing the commands from the controller view operation. Its advantage to buffer and manage commands over channels is that it is easy to extend the instructions and can be real-time and efficient at the language level, runtime level, support for parallel input, and is easy to apply to existing projects.

Abstract:This study discusses the secondary development of the database of mechanisms based on the Creo platform, and introduces the key technologies in detail. In the environment of Visual Studio 2010, with the secondary development tools Creo/TOOLKIT provided by Creo software, combined with SQL Server database, it uses MFC technology to design and compile the database of mechanisms. All mechanisms are designed by top-down in the database of mechanism. Taking the design of planetary gear cam mechanism of closed planetary gear train as an example, the specific functions of the database of mechanisms are introduced. The engineering application shows that the system can retrieve mechanism fast, and can view the mechanism's motion simulation, motion curves, features, etc. At the same time, it can design the concrete mechanism rapidly, improve the level of product design, and shorten the design cycle.

Abstract:In order to accurately predict the cost of construction project, according to the sample time dependent and nonlinear characteristics, this study has constructed a prediction model of construction project cost based on the geostatistics and support vector machine. Firstly, the cost data of construction engineering are collected, and then, the embedding dimension of time series is obtained by geostatistics according to time correlation of construction cost sample, and the construction cost learning samples of construction project, support vector machine is used to establish project cost prediction model, and test and analysis of the performance through the prediction example. The results show that the proposed model can effectively fit the construction cost of sample time correlation, and get higher accuracy of construction cost prediction, and the result is much better than the other models.