Neighborhood preserving embedding (NPE) is a classical and very promising supervised dimensional reduction (DR) technique based on a linear graph, which preserves the local neighborhood relations of the data points. However, NPE uses the K nearest neighbor (KNN) criteria for constructing an adjacent graph which makes it more sensitive to neighborhood size. In this article, we propose a novel DR method called weighted neighborhood preserving ensemble embedding (WNPEE). Unlike NPE, the proposed WNPEE constructs an ensemble of adjacent graphs with the number of nearest neighbors varying. With this graph ensemble building, WNPEE can obtain the low-dimensional projections with optimal embedded graph pursuing in a joint optimization manner. WNPEE can be applied in many machine learning fields, such as object recognition, data classification, signal processing, text categorization, and various deep learning tasks. Extensive experiments on Olivetti Research Laboratory (ORL), Georgia Tech, Carnegie Mellon University-Pose and Illumination Images (CMU PIE) and Yale, four face databases demonstrate that WNPEE achieves a competitive and better recognition rate than NPE and other comparative DR methods. Additionally, the proposed WNPEE achieves much lower sensitivity to the neighborhood size parameter as compared to the traditional NPE method while preserving more of the local manifold structure of the high-dimensional data.

No abstract

In order to study all the advantages and disadvantages of digital media enterprises, technological innovation can only be completed through cooperation. A kind of industry-university-research cooperative innovation evolutionary game method based on GS algorithm is proposed for digital media enterprise clusters. This method analyzes the evolutionary game theory of innovation and puts forward the evolutionary stability strategy of cooperative innovation between enterprises and research institutions. The results show that decreasing V is beneficial for the evolutionary game to approach the equilibrium point (1,1); that is, the greater the cost of independent innovation is compared with collaborative innovation, the stronger the willingness of both sides of the game to collaborative innovation. Enterprises and scientific research institutions are two different subjects with different interests. If they want to complete innovation cooperation, they need to formulate a perfect set of rules so that both sides of the game can carry out cooperative innovation according to the principles, so as to achieve the goal of cooperation.

Huang & Chen (2017) identified two male specimens from Nanling, North Guangdong as Lucanus brivioi ssp. incert. In the present paper, we describe and illustrate it as a new species L. zhuxiangi sp. n., basing on a large series of male and female specimens from Hunan and Guangdong, southeast China. Its diagnosis from congener species is provided. And the new species is compared with its most similar species L. brivioi Zilioli, 2003, on some selected but important morphological characters which are illustrated with color plates. MATERIALS AND METHODSSpecimens were relaxed and softened in hot water for 24 hours, and then transferred to distilled water to clean, dissect and observe. In order to examine the genitalia of both sexes, the last two abdominal segments were detached and treated with 10% solution of potassium hydroxide for 12 hours and then preserved in 75% ethanol. After examination, the body parts were mounted on a glass slide with Euparal Mounting Medium for future studies. Habitus photographs were taken using a Canon macro photo lens EF-S 60mm on a Canon 5DsR. Photographs of morphological details were taken using a Zeiss Axio Zoom.V16 motorized stereo zoom microscope with a Zeiss AxioCam MRc 5. The final deep focus images were created with Zerene Stacker 1.04 stacking software. Adobe Photoshop CS6 was used for post-processing. The morphological terminology follows Huang & Chen (2010, 2013.

No abstract