With the increase in the impact of social networking in our daily lives, the world around us has become more transparent and dynamic. Our research is focused on bringing such technological benefits to social media users in the form of useful updates. These updates will make online users aware of sentiments in community gossip, the spread of global media trends, and responses to their posts on social media. For this purpose, we have conducted a case study over a social media trend on Twitter and proposed a set of analytical metrics that can assist Twitter developers in making their social platform more informative and interactive. Though presently Twitter offers certain basic parameters through API calls, the user interface lacks insightful statistics for the benefits of the end-users. Our case study is conducted in two phases, involving basic and advanced social network analysis. The basic research includes finding users who triggered the online trend, featuring the popularity of the trend, acquiring the location of users who mentioned the hashtag, and detecting the devices that have been used frequently to tweet on the trend. We proposed to the developers advanced solutions with more insightful analytics, which involves the identification of users whose content received maximum reach and reveals influential users’ popularity amongst others tweeting with the same hashtag. Furthermore, the actual and potential reach of online social trends is computed along with sentiment analysis. Our work also suggests the availability of an ego-centric network of the most influential user in the trend for visualizing real-time diffusion of its reach.

To ensure and guarantee the reliability of lifting structures and equipment, mechanics need effective control tools to determine the condition and lifetime of these structures. This study is designed to estimate the reliability and damage of the wire rope from the behavior of its components (wires, strand, layers) and to find an analytical model that generalizes the behavior of the whole wire rope. It is also designed to establish the relationship between damage and reliability in the case of compound systems.

In order to address the drawbacks of recognition algorithm used for recursive system convolutional codes (RSC), such as low fault tolerance and high computational complexity, a novel recognition algorithm was proposed in this study. Firstly, it clearly demonstrated that the rational fraction in binary domain is capable of expanding into recurring series, and the problem of cyclic period could be solved by means of impulse response and analytical matrix. Secondly, in order to reduce the workload of computation placed on the algorithm, a polynomial database was constructed by traversing the constructed RSC code. Then, the specific matrix operation was conducted. In case of a correct ergodic polynomial, the result vector code weight tends to be significantly larger than the ergodic polynomial is incorrect, so as to realize the recognition of the polynomial. Finally, as revealed by the theoretical analysis, the fault-tolerance of the proposed algorithm was solely relevant to the code weight of the recurring series rather than the coding constraint length. The simulation results validated not only the effectiveness of the algorithm but also the correctness of the fault-tolerant performance analysis. When the error code reached as high as 0.1, the recognition rate of some polynomials was higher, and the computational complexity was lower compared to the existing algorithms.

Network security situation prediction has been a major research focus in the field of network security in recent years. It can predict the future network security status and its changing trends based on existing network security data to provide guidance for network security administrators' selection of security strategies. In this paper, a network security situation prediction algorithm based on BP neural network optimized by SA-SOA is proposed. The algorithm uses the seeker optimization algorithm (SOA) to find the best fitness individual, obtains the optimal weight and threshold value, assigns them to the random initial threshold value and weight value of the BP neural network, and finally obtains the prediction value through the training of the BP neural network. To solve the problem that the seeker optimization algorithm is easy to fall into the local optimization and slow convergence in the later stage of the search, the simulated annealing algorithm (SA) is introduced into the seeker optimization algorithm. According to the Metropolis criterion of SA, the algorithm accepts the bad solution with a certain probability, which avoids falling into the trap of the local optimum and improves the global search ability of the algorithm. The experimental results show that this algorithm is more accurate and more stable than other prediction algorithms based on the improved BP neural network.

Breast cancer is the most common cancer that affects women in the world, and predicting breast cancer viability is a complex and challenging work. Most previous efforts focused on statistical or supervised approaches to predict the survival prospects of patients. However, both complex feature correlation and missing feature values may increase the difficulty of survivability prediction. In this paper, we propose a novel hybrid model to ameliorate patient survivability analysis. Firstly, we utilize principal component analysis (PCA) and K-means to create clusters with similar characteristics based on the different stages of breast cancer spread. Then, these clusters are exploited to train the classification model for patient survivability prediction. Experimental results show that, compared with the original historical data, the accuracy of survivability prediction for specific models is further improved by using identified patient cohorts.

Partial discharge (PD) is used in high-voltage cable online monitoring, the collected signal contains a variety of noises, usually periodic narrow-band interference and white noise are the most common and the most widely affected. In order to suppress the influence of complex noise, a partial discharge signal noise reduction method based on Variational Mode Decomposition (VMD) is proposed. VMD is used to decompose the noisy partial discharge signal to obtain the intrinsic mode functions (IMF) from low to high frequency, then calculate the kurtosis value of each variational modal component, the pulse characteristic component for reconstruction was select, and adaptive threshold used to reduces the noise of the reconstructed signal again. Compared with the wavelet transform threshold method, the noise reduction results are obtained in different noise environments. The results are quantitatively superior to the wavelet standard soft threshold noise reduction method and the wavelet global hard threshold noise reduction method from the mean square error and the waveform similarity coefficient. Simulation and field experiment results show that this method can effectively remove noise signals, and can retain the original signal waveform more completely, which is convenient for subsequent signal processing.

Traffic light and traffic sign detection are important in the field of self-driving. They can guide vehicles to drive safely on the road. It is difficult for existing algorithms of object detection to detect targets simultaneously and achieve high accuracy. In this paper, a dual-model framework is proposed to detect traffic light and signs for a self-driving vehicle based on prior information. This framework can switch the detection model according to the prior information. The color information of the traffic sign is used to extract the ROI and improve the detection efficiency. The work of this paper also includes collecting and annotating a large amount of image data to apply the model trained on the proposed framework to self-driving. The proposed framework is verified on a real road test of a self-driving vehicle.

This paper investigates the problem of moving target tracking in the case of a monocular mobile camera. Firstly, a scene space model was established under a moving camera based on the constraint conditions of ground plane. Thus, a three-dimensional (3D) motion model of target could be established. The scene space model and the target 3D motion model could both be compensated by the posture of the camera, which can be calculated according to the vanishing point in the image or the motion parameters of the camera acquired by sensors. Secondly, an interception function was established to extract the target motion range according to the target 3D motion model. Both the influence of background factors and the computational complexity of the algorithm were reduced. Then, a dual-kernels tracking method was used to track the target under a situation of non-obvious color characteristics and partial shielding. Finally, a public data set was used to conduct comparison experiments with three other algorithms to verify the effectiveness and robustness of the proposed algorithm.

Due to the high dimensions of complex hyperspectral data features and the difficulty of feature selection and extraction, it is difficult to construction an inversion model. This paper briefly describes the common methods of hyperspectral data analysis and processing. To address the existing problems in the field of hyperspectral data analysis, an integrated deep learning framework combined with artificial neural network and large GAN is proposed in this paper. The application of this integrated deep learning framework in hyperspectral data analysis is discussed further. Results show that the integrated deep learning method combining neural network and large GAN provides direction for hyperspectral data analysis and processing, which has a broad application prospect.

A heterogeneous network is an environment with different types of interconnected networks that contains a variety of communication links, networking modes, and communication protocols. The heterogeneity and variability of such networks have triggered a range of problems, such as security risks, unreasonable communication-resource allocation, the disordering of network access, and difficulties in ensuring quality-of-service (QoS) with business priorities. In this context, a QoS routing solution is proposed to combine the weighted-fair-queuing (WFQ) algorithm and an extended open-shortest-path-first (OSPF) protocol. The network model, multi-service transmission, and routing protocol of a marine-oriented heterogeneous network are simulated by means of OPNET. The proposed QoS routing solution is then verified in a high-load simulation environment. The experimental results prove that the solution can improve network resource control and routing optimization to some extent by reducing such QoS parameters as transmission delay of different services, average delay, and jitter of network.

Avionics equipment is playing an increasingly important role in the integration of avionics systems. Due to the harsh airborne environment, such as high and low temperature, strong impact, acid corrosion and other factors, the fault features of avionics equipment have strong randomness, complexity, and coupling with external stress, which leads to the support technology of avionics equipment becoming a research hotspot. Aiming at the classical problems of engineering, this paper proposes a fault prognosis method based on the fusion of transferred multi-deep-belief network (DBN) models. First, Transfer learning and Dropout strategy are used to improve the feature extraction capability of the model. Secondly, the optimized genetic algorithm effectively determines the fusion weight of each DBN model according to the real-time data. Finally, the complete hybrid framework was adopted to estimate the complete residual life of the equipment. To verify the performance of the proposed method, the experiment was performed according to the data of the power supply equipment. The results show that the proposed method has higher accuracy and stability than traditional support vector machines and classical DBN models which is helpful to realize the maintenance based condition for avionics equipment.

Motion target detection plays an important role in object tracking and traffic monitoring fields. To improve the estimation accuracy of motion direction and thereby increase restoration resolution, research of blurred image spectrum analysis, blurred direction detection, and interference suppression was conducted. Firstly, the interference of bright cross in the spectrum caused by the edge truncation effect was eliminated through by partitioning the spectrum image. Then, the diagonal interference was reduced by normalized Radon transform. Finally, to improve the estimation accuracy of the blurred direction of any ration image, the factor aspect ratio was used to modify the relation between the tilt angle of the dark stripe and the motion direction angle. The experiment demonstrates that this algorithm can effectively avoid interference peaks appearing in the 0°, 90°, and 45° (135°) directions of the blurred direction estimation curve, and it can also accurately estimate the blurred direction in the non-diagonal with an error of about 4°-6°. Overall, this algorithm can suppress the interferences of bright cross and diagonal effectively and can be applied to any aspect ratio.

Cloud computing with high availability and good scalability is becoming an important platform to solve scientific workflow problems, and scheduling is the prominent issue for optimal strategy. This paper presents a cloud scheduling algorithm using graph partitioning and reinforcement learning for data-intensive workflow scheduling based on the cloud heterogeneous platform. Aiming at reducing the cost and makespan during the process of task execution, the proposed algorithm firstly evaluates the dual optimization weight of workflow and clusters the tasks that have strong data dependence into blocks by the graph partitioning algorithm. Then, we train the partitioned workflow by iterating the reinforcement learning method such that the match for tasks and resources will meet our expectations. In consideration of performance metrics like total cost and makespan, the partitioning reinforcement learning scheduling algorithm is much better compared to classic algorithms during the simulating experiments conducted on the scientific workflow simulation.

The task of font effect generation is stylizing the shape and texture of style images into font images. There exist some methods to handle this task. However, stylized font images become unrecognized when the glyph structure is quite complicated. This paper proposes a font effect generation model based on pyramid style feature. Morphology operations are utilized to improve the transferring effect. Experiments show that our proposed method is more suitable for stylizing complex glyph images than other state-of-the-arts methods.

It has been proved by practice that mining weblogs to detect software errors is an effective software testing method. This paper presents a software failure detection method based on the incremental mining weblog strategy and gives data mining steps for the implementation of this method. A case is studied for the proposed test method. In this case, we use Splunk, a data analysis tool, to analyze some weblogs that record some linked information of mobile applications for android. The result of the data analysis shows that the proposed method can effectively detect the software failure problem in the download process of mobile applications. Therefore, the proposed method can be used for software reliability assessment.

Software security bug is one of the key threats to the security of software systems. Isolating security bugs that may be potential security bugs is important. We formalize a program error propagation based model (PEP), which used to be applied to locate integer bug and our contribution are as follows: We formulate a theory model based on the mechanism of how the security bug triggers the program error propagation and propose a security bug localization approach by applying spectrum-based fault-localization (SFL) technique, a novel method to locate software fault to alleviate false negative and false positive problem. Our experimental results show that：1)Our model is more effective than present ones to locate nearly 97% integer bug and buffer overflow which are the main security bugs by examining 50% codes on average; 2) Compared with the traditional techniques, SFL can find 100% of integer bugs and buffer overflow so it is a promising, technology roadmap to reduce false negative and false positive for locating security bugs.

In the development of intelligent architecture industry, the issue of building energy conservation has been regarded as one of the most urgent problems. As a scientific and technological product, the photovoltaic (PV) system has brought building energy-saving technology into a new era. At the same time, it combines architecture and art very well. The building integrated PV (BIPV) industry will be the commanding height of the competition of the world's advanced industry in the future. Thus, it is significant to study the photovoltaic system in intelligent photovoltaic architecture. In this paper, we introduce the principle of the PV system and discuss the creativity of the PV system in architecture. Finally, we present a creative design of a set of grid-connected PV system and analyse the benefit of the system. From the research, we can draw a conclusion that the intelligent photovoltaic architecture is not only an excellent original architectural art but also an energy conservative building. It will be beneficial for the study of the PV system in the field of building energy efficiency and architectural art.

Aiming to precisely detect targets in complex scenes using optical remote sensing technology, the technology of modeling and simulation for polarimetric detection on remotely-sensed targets is discussed. Due to the complex interference between targets, background reflection, and aerosol radiation in real environments, the modeling and simulation of polarimetric targets are difficult. Firstly, the optical theories and basics of polarimetric model are introduced, and then a numerical simulation and analysis for Priest-Germer model are given. The Priest-Germer model obtains different results for different materials and is capable of distinguishing materials. The design idea and software frame of polarimetric detection software are provided. Finally, giving two kinds of material and two wavelength irradiation (440nm, 600nm), four groups of simulation experiments for two targets and polarimetric image histogram comparison are exhibited. The derived results prove that polarimetric simulation software based on PG model is sensitive to different targets, irradiation, and materials and provides distinguishing ability. The visualization effects are relatively good, the software has great values on practice, and it is shown that the target model, atmospheric transmission model, and physical property of materials have comprehensive and decisive effects on polarimetric imaging.