As a crime of employing technical means to steal sensitive information of users, phishing is currently a critical threat facing the Internet, and losses due to phishing are growing steadily. Feature engineering is important in phishing website detection solutions, but the accuracy of detection critically depends on prior knowledge of features. Moreover, although features extracted from different dimensions are more comprehensive, a drawback is that extracting these features requires a large amount of time. To address these limitations, we propose a multidimensional feature phishing detection approach based on a fast detection method by using deep learning. In the first step, character sequence features of the given URL are extracted and used for quick classification by deep learning, and this step does not require thirdparty assistance or any prior knowledge about phishing. In the second step, we combine URL statistical features, webpage code features, webpage text features, and the quick classification result of deep learning into multidimensional features. The approach can reduce the detection time for setting a threshold. Testing on a dataset containing millions of phishing URLs and legitimate URLs, the accuracy reaches 98.99%, and the false positive rate is only 0.59%. By reasonably adjusting the threshold, the experimental results show that the detection efficiency can be improved. INDEX TERMS Phishing website detection, convolutional neural network, long short-term memory network, semantic feature, machine learning.

No abstract

No abstract

No abstract

No abstract

Magnetic bearings have no mechanical contact between the rotor and stator, and a rotary pump with magnetic bearings therefore has no mechanical wear and thrombosis. The magnetic bearings available, however, contain electromagnets, are complicated to control and have high energy consumption. Therefore, it is difficult to apply an electromagnetic bearing to a rotary pump without disturbing its simplicity, reliability and ability to be implanted. The authors have developed a levitated impeller pump using only permanent magnets. The rotor is supported by permanent radial magnetic forces. The impeller is fixed on one side of the rotor; on the other side the rotor magnets are mounted. Opposite these rotor magents, a driving magnet is fastened to the motor axis. Thereafter, the motor drives the rotor via magnetic coupling. In laboratory tests with saline, where the rotor is still or rotates at under 4,000 rpm, the rotor magnets have one point in contact axially with a spacer between the rotor magnets and the driving magnets. The contacting point is located in the center of the rotor. As the rotating speed increases gradually to more than 4000 rpm, the rotor will disaffiliate from the stator axially, and become fully levitated. Since the axial levitation is produced by hydraulic force and the rotor magnets have a giro-effect, the rotor rotates very stably during levitation. As a left ventricular assist device, the pump works in a rotating speed range of 5,000-8,000 rpm, and the levitation of the impeller is assured by use of the pump. The permanent maglev impeller pump retains the advantages of the rotary pump but overcomes the disadvantages of the leviated pump with electromagnetic-bearing, and has met with most requirements of artificial heart blood pumps, thus promising to have more applications than previously.

An increasing number of grounded robots are being used in prostate interventions to improve clinical outcomes, but their large size and high-cost limit their popularity. Thus, we present a hand-held 3-degree of freedom (DoF) parallel robot with a-remote center of motion (RCM) for minimally invasive prostate biopsy applications, combining the flexibility of hand-held devices with the precision of robotic assistance. First, the kinematic structure of robotic assistance is introduced according to its design requirements. Then, the kinematic analysis of robotic assistance is carried out by using a simplified kinematic model. The kinematic parameters are designed according to the desired workspace. A prototype has been developed and validated in animal experiments. Twenty beagles of different sizes were selected for the robot-assisted and controlled experiments, resulting in target errors of 3.30 ± 1.63 mm and 5.40 ± 1.76 mm, respectively. The error of robot-assisted experiments was significantly better than in controlled experiments. Preliminary animal tests have demonstrated that the hand-held robot can improve the accuracy of free-hand biopsy punctures.

To optimize the energy efficiency of edge computing system with energy harvesting, this paper proposes an energy-efficient task offloading method optimized by differential evolution. First, a wireless edge computing network model is established to analyze the energy harvesting, task offloading and task calculation of the system, as well as the total number of calculated bits and total energy consumption of the system. Second, according to the total number of calculated bits and total energy consumption of the system, an objective function is established to optimize the energy efficiency of system, and a differential evolution based optimization method is proposed, with which the optimal energy efficiency of system calculation, offloading time, calculation time and frequency are obtained. Experimental results show that the proposed method can not only achieve better convergence effect, but also can effectively solve the energy shortage problem of the micro-equipment and extend the service life of the equipment.INDEX TERMS Edge computing, task offloading, energy harvesting, differential evolutionary algorithm.PENG ZENG received the Ph.D. degree in mechatronic engineering from the Graduate School of the