Zheng X.,Fuzhou University |
Zheng X.,A+ Network |
Zeng Z.,Fuzhou University |
Zeng Z.,A+ Network |
And 4 more authors.
Neurocomputing | Year: 2015
Social network has become a very popular way for internet users to communicate and interact online. Users spend plenty of time on famous social networks (e.g., Facebook, Twitter, Sina Weibo, etc.), reading news, discussing events and posting messages. Unfortunately, this popularity also attracts a significant amount of spammers who continuously expose malicious behavior (e.g., post messages containing commercial URLs, following a larger amount of users, etc.), leading to great misunderstanding and inconvenience on users' social activities. In this paper, a supervised machine learning based solution is proposed for an effective spammer detection. The main procedure of the work is: first, collect a dataset from Sina Weibo including 30,116 users and more than 16 million messages. Then, construct a labeled dataset of users and manually classify users into spammers and non-spammers. Afterwards, extract a set of feature from message content and users' social behavior, and apply into SVM (Support Vector Machines) based spammer detection algorithm. The experiment shows that the proposed solution is capable to provide excellent performance with true positive rate of spammers and non-spammers reaching 99.1% and 99.9% respectively. © 2015 The Authors.
Chen P.,QingHua University |
Chen P.,Chongqing University of Posts and Telecommunications |
Wu G.,Chongqing University of Posts and Telecommunications |
Lu S.,Chongqing University of Posts and Telecommunications
BioTechnology: An Indian Journal | Year: 2013
With the increasingly serious environmental pollution and energy crisis, the new energy automotive industry has brought a new opportunity for development. With the development of DC/DC power module to the direction of high power density as the important parts of the new energy vehicles, thermal reliability research becomes increasingly important. The thesis used the finite element method to make the thermal analysis in order to solve the the thermal reliability problems of the DC/DC power module, then temperature distribution over the entire module was obtained; it verified the validity of finite element thermal analysis model compared with the experimental data. At last thermal design methods based on the analysis were proposed. It proved The highest temperature of the DC /DC power module dropped from 70. 6 °C to 39. 7 °C and the temperature range became 9.5°C from 43.3°C after the thermal design optimization by using the finite element thermal analysis methods. The result shows that DC / DC power module thermal reliability is improved by he thermal design optimization. © 2013 Trade Science Inc.
Cao G.,South China University of Technology |
Luo P.,QingHua University |
Luo P.,GRG Banking Equipment Co. |
Wang L.,Hebei University of Technology |
Yang X.,Hebei University of Technology
Procedia CIRP | Year: 2016
With the rapid development of global economy, sustainable problems have aroused widespread concern by all countries in the world. Sustainable design is a systematic design method to develop the sustainable products or form the solutions to sustainable problems to adapt the sustainable development. This article was a study of key technology for sustainable design through the patent knowledge mining, aiming to provide technical support and process model to the transformation from traditional products to sustainable products. On the one hand, sustainable design criteria for design goal, life cycle and design strategy are extracted and established through the patent knowledge mining and the product sustainable requirements. On the other hand, the laws of function organization and improvement are extracted by patent knowledge mining and dynamic analysis of function evolution. Then the process model of sustainable product function innovation was put forward by integrating sustainable design criteria, functional evolution laws and TRIZ into traditional function design. Finally, the buckle ring machine was taken as an example to validate the effectiveness of the model. © 2016 The Authors.
Du J.,Yangzhou University |
Huang Y.-G.,Peking Union Medical College |
Yu X.-R.,Peking Union Medical College |
Zhao N.,QingHua University
Chinese Medical Journal | Year: 2011
Background Ketamine is hypothesized to reduce perioperative endocrine-metabolic and inflammatory responses in cardiac surgery patients. This randomized, placebo-controlled, double-blind study was performed to determine whether perioperative endocrine-metabolic and inflammatory responses are attenuated by preoperative administration of ketamine to healthy females receiving elective laparoscopic surgery. Methods Forty female patients with American Society of Anesthesiologist classification I or II who elected to receive gynecological laparoscopic surgery were randomly assigned to the ketamine-treated (group K; n = 20) or control (group C; n = 20) group. At 2 minutes prior to induction patients in group K received ketamine (0.25 mg/kg) whereas those in group C received normal saline. All patients received standardized general anesthesia. Serum glucose and cortisol values were measured before ketamine administration (T0), 2 minutes after tracheal intubation (T1), 30 minutes after skin incision (T2), 2 minutes after tracheal extubation (T3) and 1 hour postoperatively (T4). Serum interleukin-6 and tumor necrosis factor-α values were determined at T0 and T4. Postoperative analgesic efficacy, side effects of administered drugs, and time to discharge were recorded. Results Compared with subjects in group C, those in group K had lower serum glucose values at T1, T2, T3 and T4 and lower serum cortisol values at T4 (P <0.05). Postoperative interleukin-6 and tumor necrosis factor-α concentrations for group K were lower than those for group C (P <0.05). Postoperative visual analog scale scores at rest, cumulative fentanyl consumption, and time to discharge were lower in group K as compared to group C (P <0.05). No significant differences in drug side effects were observed postoperatively between the two groups. Conclusion Endocrine-metabolic and inflammatory responses to laparoscopic surgery are attenuated in part by pre-incisional administration of ketamine.
Chen D.,Xian Jiaotong University |
Wang Y.,Xian Jiaotong University |
Bai Y.,Xian Jiaotong University |
Wang Y.-H.,QingHua University |
And 4 more authors.
Wuji Cailiao Xuebao/Journal of Inorganic Materials | Year: 2015
Effect of Reynolds number on the flattening behavior of in-flight particles was investigated by slit method. Molten 6wt%-8wt% yttria-stabilized zirconia (YSZ) particles with different Reynolds numbers were collected. The morphology of splats was examined using scanning electron microscope (SEM) and 3D laser microscope. Meanwhile, the behavior of spreading, heat transfer and solidification of YSZ droplets with different Reynolds numbers were simulated by CFD software Fluent. The results show that the flattening behavior of molten particles is significantly influenced by Reynolds number when Ohnesorge number exceeds 0.2. The splat tends to splash with the increase of Reynolds number. At a lower Reynolds number, the splats exhibit a regular disc-shape. However, when Reynolds number reaches 450±20, splats begin to splash. In addition, for the original molten particles with the same diameters, the flattening ratio of molten particles in supersonic atmospheric plasma spraying (SAPS) is approximately 1.3 times as much as that in atmospheric plasma spraying (APS). ©, 2015, Science Press. All right reserved.