Unsupervised cross-modal hashing has achieved great success in various information retrieval applications owing to its efficient storage usage and fast retrieval speed. Recent studies have primarily focused on training the hash-encoded networks by calculating a sample-based similarity matrix to improve the retrieval performance. However, there are two issues remain to solve: (1) The current sample-based similarity matrix only considers the similarity between image-text pairs, ignoring the different information densities of each modality, which may introduce additional noise and fail to mine key information for retrieval; (2) Most existing unsupervised cross-modal hashing methods only consider alignment between different modalities, while ignoring consistency between each modality, resulting in semantic conflicts. To tackle these challenges, a novel Deep High-level Concept-mining Jointing Hashing (DHCJH) model for unsupervised cross-modal retrieval is proposed in this study. DHCJH is able to capture the essential high-level semantic information from image modalities and integrate into the text modalities to improve the accuracy of guidance information. Additionally, a new hashing loss with a regularization term is introduced to avoid the cross-modal semantic collision and false positive pairs problems. To validate the proposed method, extensive comparison experiments on benchmark datasets are conducted. Experimental findings reveal that DHCJH achieves superior performance in both accuracy and efficiency. The code of DHCJH is available at Github.

Ground-based cloud image sequence prediction provides valuable insights into cloud motion and meteorological conditions, which are essential for photovoltaic power generation systems. Most existing models are, however, recurrent-based, which is problematic in providing satisfactory forecasting results with rapid speed because these recurrent-based models do not support parallel inference and usually suffer from slow inference speed. A novel recurrent-free deep learning-based framework, called multiscale spatiotemporal attention network (MSTANet) to address the issues is proposed in this study. The MSTANet leverages a multiscale spatiotemporal attention (MSTA) module to extract the multiscale, nonlinear spatiotemporal dependencies from cloud image sequences and uses a multiscale temporal attention (MTA) module to reinforce the temporal dependencies by capturing the high- and low-frequency spatiotemporal fluctuations of clouds. A gated aggregation unit (GAU) to mitigate the ghosting effects that are prevalent in spatiotemporal prediction tasks is introduced to filter the useful context information by integrating the historical information with the updated predictions. Additionally, a multiorder differential divergence regularization term is introduced into the loss function to improve the model’s performance by encouraging MSTANet to focus on the evolving trends of the neighborhood of clouds. Experimental results show that the proposed MSTANet outperforms the state-of-the-art (SOTA) prediction methods. It reduces 46% parameters and mean-squared-error (MSE) by 4.31% on the Moving Mnist dataset and reduces 22% parameters with a 1.82% performance improvement on the Folsom dataset compared to the baseline temporal attention unit (TAU). The codes are available athttps://github.com/Csorasky/MSTANet.

Stanley algorithm is a classical algorithm in automatic driving path tracking algorithm, which has the advantages of low computational complexity and good tracking effect. However, the tracking accuracy of the traditional Stanley algorithm is limited by the gain parameters and the gain coefficients cannot be dynamically adjusted according to the road conditions. To improve the tracking accuracy and driving stability, an adaptive Stanley control algorithm based on DWA (DWA-Stanley) is proposed, which is used to sampling the velocity vector space and selecting the evaluation function, the vehicle motion planning is based on the best set of velocity values. In this paper, the DWA-Stanley algorithm is validated based on three different driving speeds. Simulation results show that an racecar using the DWA-Stanley algorithm has higher tracking accuracy and smoothness than a conventional Stanley.

Path tracking is an important part of autonomous vehicles. Stanley algorithm is widely used in path tracking control of front wheel steering vehicles. The traditional Stanley algorithm calculates the front wheel angle according to the relative geometric relationship between the vehicle pose and the reference path point, which depends on the nearest reference path point. It is suitable for the low-speed and small change in curvature of reference paths. In order to improve the tracking accuracy and stability of Stanley algorithm, a model predictive control path tracking algorithm based on adaptive Stanley is proposed, which considers the adaptive change of preview distance and vehicle dynamics. The comparative simulation analysis of the proposed control strategy and the traditional Stanley control shows that the model predictive control path tracking algorithm based on adaptive Stanley can maintain high trajectory tracking accuracy and vehicle stability at high speed.

The requirement of higher education is increasing in China caused by the development of economics and growing population. Many of universities in China enlarge their campus to meet the demand of abundant students which led longer commutes between two areas in campus. Campus bus using an environmental friendly way to increase the commute efficiency which have important significance to realize carbon reduction target. This paper introduces a method to design environmental friendly campus bus route taking Shenzhen Technology University (SZTU) as an example.

Influenced by many complex factors, it is very difficult to obtain high-performance industrial power load forecasting. The industrial power load forecasting is deeply studied by fusing some machine-learning methods for industrial enterprise power consumers. As a result, a novel power load forecasting method is proposed by taking into account the variation of load characteristics in different regions, industries, and production patterns. First, through the improved K -means clustering analysis, the historical load data are classified as the production patterns to which they belong. Then, the prediction algorithm combining reinforcement learning with particle swarm optimization and the least-squares support vector machine is proposed. Finally, the improved algorithm in this article is used for short-term load forecasting separately by the load data in different patterns after the above processing. The forecasting method in this article is based on data driven with real datasets. The results of the simulation experiment show that the improved prediction algorithm can distinguish the changes in different production patterns and identify the load characteristics of different regions and industries with high prediction accuracy, which has practical application value.

Genome-scale metabolic network models (GEMs) have played important roles in the design of genetically engineered strains and helped biologists to decipher metabolism. However, due to the complex gene-reaction relationships that exist in model systems, most algorithms have limited capabilities with respect to directly predicting accurate genetic design for metabolic engineering. In particular, methods that predict reaction knockout strategies leading to overproduction are often impractical in terms of gene manipulations. Recently, we proposed a method named logical transformation of model (LTM) to simplify the gene-reaction associations by introducing intermediate pseudo reactions, which makes it possible to generate genetic design. Here, we propose an alternative method to relieve researchers from deciphering complex gene-reactions by adding pseudo gene controlling reactions. In comparison to LTM, this new method introduces fewer pseudo reactions and generates a much smaller model system named as gModel. We showed that gModel allows two seldom reported applications: identification of minimal genomes and design of minimal cell factories within a modified OptKnock framework. In addition, gModel could be used to integrate expression data directly and improve the performance of the E-Fmin method for predicting fluxes. In conclusion, the model transformation procedure will facilitate genetic research based on GEMs, extending their applications.

For realizing non-intrusive protection of open SCADA systems, a non-intrusive solution for distributed open SCADA systems is proposed. The solution consists of three functionality parts: Abstract Execution, Refine State, and Behavior Checking. The approach provides a runtime verification of the system by combining cyclic semantic reconstruction of VM and abstract execution of SCADA services. First, all Internet packets through virtual network bridges are extracted and symbolically linked to specific service model to get simulated traces. Then, cyclic semantic reconstruction is performed to acquire the current service runtime state. According to the service instance state of semantic reconstruction, the simulated traces are refined. When a trace is identified, behavior checking is adopted to verify whether the runtime state is compliant to the system specification that is defined based on milestone events for meeting SCADA real-time requirements.

MOTIVATION: In recent years, genome-scale metabolic models (GEMs) have played important roles in areas like systems biology and bioinformatics. However, because of the complexity of gene-reaction associations, GEMs often have limitations in gene level analysis and related applications. Hence, the existing methods were mainly focused on applications and analysis of reactions and metabolites. RESULTS: Here, we propose a framework named logic transformation of model (LTM) that is able to simplify the gene-reaction associations and enables integration with other developed methods for gene level applications. We show that the transformed GEMs have increased reaction and metabolite number as well as degree of freedom in flux balance analysis, but the gene-reaction associations and the main features of flux distributions remain constant. In addition, we develop two methods, OptGeneKnock and FastGeneSL by combining LTM with previously developed reaction-based methods. We show that the FastGeneSL outperforms exhaustive search. Finally, we demonstrate the use of the developed methods in two different case studies. We could design fast genetic intervention strategies for targeted overproduction of biochemicals and identify double and triple synthetic lethal gene sets for inhibition of hepatocellular carcinoma tumor growth through the use of OptGeneKnock and FastGeneSL, respectively. AVAILABILITY AND IMPLEMENTATION: Source code implemented in MATLAB, RAVEN toolbox and COBRA toolbox, is public available at https://sourceforge.net/projects/logictransformationofmodel.

We investigate essential relationships between generalization capabilities and fuzziness of fuzzy classifiers (viz., the classifiers whose outputs are vectors of membership grades of a pattern to the individual classes). The study makes a claim and offers sound evidence behind the observation that higher fuzziness of a fuzzy classifier may imply better generalization aspects of the classifier, especially for classification data exhibiting complex boundaries. This observation is not intuitive with a commonly accepted position in “traditional” pattern recognition. The relationship that obeys the conditional maximum entropy principle is experimentally confirmed. Furthermore, the relationship can be explained by the fact that samples located close to classification boundaries are more difficult to be correctly classified than the samples positioned far from the boundaries. This relationship is expected to provide some guidelines as to the improvement of generalization aspects of fuzzy classifiers.

Border Gateway Protocol (BGP) is the core component of the Internet's routing infrastructure. Abnormal routing behavior impairs global Internet connectivity and stability. Hence, designing and implementing anomaly detection algorithms is important for improving performance of routing protocols. While various machine learning techniques may be employed to detect BGP anomalies, their performance strongly depends on the employed learning algorithms. These techniques have multiple variants that often work well for detecting a particular anomaly. In this paper, we use the decision tree and fuzzy rough set methods for feature selection. Decision tree and extreme learning machine classification techniques are then used to maximize the accuracy of detecting BGP anomalies. The proposed techniques are tested using Internet traffic traces.

This paper proposes a method to fuse Real-valued K nearest neighbor classifier by feature grouping. Real-valued K nearest neighbor classifier can approximate continuous-valued target functions, which can provide more information than crisp K nearest neighbor classifier in fusion. In addition real-valued K nearest neighbor classifier is sensitive to feature perturbation. Therefore, when multiple real-valued K nearest neighbor classifiers are fused by feature grouping, the performance of the fusion is better than single classifier. In order to validate the performance of fusion, four datasets are selected from UCI Repository. Experimental results show that the performance of fusion is better than single classifier and multiple classifier system by other perturbations.