Search

Human-oriented and machine-generated data in cyber-physical-social systems are often complicated graph-structured. Graph-powered learning methods are conducive to discovering valuable knowledge from large scale graph data and improving decision making processes. However, due to the neglect of diverse relations among things, most existing knowledge reasoning studies are inherently flawed and inefficient in processing the heterogeneous graphs with high-order connectivity. Tensor, as a powerful and effective tool to model high-level semantic interactions between various things, can provide high-order IoT graph with new perspectives and possibilities. Therefore, this paper innovatively proposes a collaborative Artificial Intelligence of Things data analysis and application framework based on Boolean tensors, which supports the expression and fusion of heterogeneous graph and ultimately promotes the AI processing. In this context, we focus on developing an incremental Boolean tensor factorization (IBTF) approach for efficient knowledge reasoning to meet the requirements of real-time and high level quality demands for intelligent services. To the best of our knowledge, we are the first to do this work. More concretely, we present factors update and binary features merge algorithms for the integrated graph tensors to avoid numerous repeated calculations of historical data. Experimental results on general synthetic data sets demonstrate that the IBTF approach proposed in this paper guarantees nearly equal approximate accuracy while reducing execution time by dozens and even more of times. Furthermore, experimental evaluations and interpretability analysis on real-world datasets verify the practicality of the proposed framework and approach.

The helix roller of the omni-directional wheel was measured by the three-coordinate measuring machine and then the obtained data was preprocessed. The principle of the least squares approximating NURBS (Non-Uniform Rational B-Spline) curve fitting algorithm was analyzed and with this algorithm the contour line of the roller was reconstructed in MATLAB software, thus the outside silhouette surface was generated on the basis of the fitting curve and finally the three-dimensional solid model of helix roller was built in Pro/Engineer. The results laid a good foundation for further finite-element analysis and manufacturing.

Sensitive text recognition has become an important technical means to maintain the harmonious development of network environment. This paper proposes a document sensitivity calculation method (SDC) based on semantic dependencies. This method associates the sensitivity of vocabulary with its semantic roles in sentences, and uses sequence annotation instead of dictionary matching. According to the semantic dependency relation, the sensitivity direction of the sentence is determined, and the sensitive vector of sentence and document level is obtained. A number of classifiers are used for comparison experiments at all levels of sensitive document sets. The experimental results show that the SVM classifier has achieved more than 80% F- measurement in all levels of documents, which is more than 10% higher than that of the control experiment. SDC algorithm reduces the propagation and diffusion of sensitive text by efficiently recognizing sensitive text.

In this paper, the underwater self-navigating platform hull was optimized calculation and analysis, in order to get minimum mass per unit volume, the optimization constraints were in addition to meet the strength and stability, also to consider the manufacturing, assembly and other factors. Based on the structure optimization, the method to determine platform hull optimal scale was given by considering of sailing resistance and flexibility of the platform comprehensively.

Neuronal morphology, which is closely related to neuronal characteristics and functions, is complex and diversified, and it has gradually attracted more and more neuroscientists to study on it. As the neuronal classification is a basic point in neuronal study but no existing universal methods take neuronal morphology into consideration, this paper dedicates to fill this blank. Firstly, this paper used Principal Component Analysis (PCA) method to select five key features from twenty neuronal morphologic features. With these key features, this paper leveraged hierarchical clustering to cluster sixty neurons randomly selected from the NeuroMorpho.Org website. As a result, these neurons were divided into four categories. Finally, we devised a new naming scheme by the range of key features' values. Experiments indicated that this classification can effectively distinguish neurons by morphology. At last, this paper discussed and analyzed the anomaly that occurs after a large number of classification experiments.

In recent years, much attention has been focused by practitioners as well as academics on a new manufacturing paradigm called 'instant customerisation' (IC). One of critical operation tactics is finalise-to-individual demand forecasting to realize IC. It attaches more importance to individual demand forecasting, especially short-term forecasts. However, there is scarcely any relevant research. According to the character of individual demand, an oriented-attribute forecasting approach is proposed. We forecast individual demand indirectly after transforming it into a time series. Then we propose a Case-based Reasoning (CBR) system for universal individual demand forecasting. The forecast process is divided into three stages: case representation, similarity search, and case forecasting and adjustment. In particular, we proposed a distance-based formula that broadens our view by regarding the slope and length of a segment as changing characteristics of a time series. The performance of our system is evaluated using real data from a restaurant. The proposed approach is compared with other traditional time series forecasting methods. Because of particularity of the data, a special error measure, MASE, is employed. The empirical results show that our method is able to produce a forecast with a high degree of accuracy.