With the rapid development of Internet technology and social networks, a large number comment texts are generated on Web. In era big data, mining emotional tendency comments through artificial intelligence is helpful for timely understanding network public opinion. The sentiment analysis part intelligence, its research very meaningful obtaining trend comments. essence text classification task, different words have contributions to classification. current studies, distributed word...

The method of text sentiment analysis based on dictionary often has the problems that doesn't contain enough words or omits some field words. In addition, due to existence polysemic with positivity, negativity, and neutrality, words' polarity cannot be accurately expressed, so accuracy is reduced extent. this paper, an extended constructed. contains basic words, which improves analysis. naive Bayesian classifier used determine in word is. Thus, value obtained. By utilizing designed score...

Abstract Many researchers have studied the application of low temperature plasma in soil remediation, but most focus on degradation efficiency a specific pollutant, and less attention is paid to development evolution process behind remediation. In this paper, possible mechanism discharge under pulse voltage from perspectives electricity optics. The electrical optical analyses show that mainly exists at rising stage voltage, existence time related with amplitude. At same time, relative...

The mining of network sensitive information is great significance for understanding the social stability network. Obtaining public opinion helpful to master Internet users' attitudes toward important events. related artificial intelligence technology can achieve topics from texts. At present, current topic recognition model has a low rate and usually generates some inaccurate keywords. In this paper, method based on word weighted-latent Dirichlet allocation (LDA) proposed. First, basic...

With the rapid development of Internet, more and users expressed their views on Internet. Therefore, big data texts are generated In era data, mining sentiment tendencies contained in massive Internet through natural language processing technology has become an important way public opinion supervision. this paper, sensitive information topics-based analysis method for is proposed. This integrates topic semantic into text representation a neural network model. The attention mechanism...

With the rapid development of Internet, number Internet users has grown rapidly, and become more influential on people's lives. As a result, amount network text is increasing it difficult to extract interested event information from only by manual reading. Therefore, extraction technique automatically extracting useful large unstructured texts becomes increasingly important. Event detection first step task plays vital role in it. However, current research lacks comprehensive consideration...

Internet is an important platform to spread public opinion for Tibetan people. The research on the Web pages content analysis meaningful monitoring. Detecting sensitive words beneficial understand of minority. In this paper, we present a novel information classification algorithm and topic tracking contents. First, text method proposed based vector space model cosine theorem. main idea different locations gives importance degrees at term weight computing. Building word list artificial work....

Electrical discharge is a sign of insulation damage power equipment and an essential part inspection. For the recognition plasma patterns, accuracy rate traditional manual inspection relies heavily on experience staff, misjudgment inevitable. The emerging method based feature extraction signals also has its limitations. Thus, high-accurate convolutional neural network (CNN) visible images proposed in this paper. Under deep learning framework-TensorFlow, CNN model established pre-trained...

Abstract Considerable research efforts have been devoted to the study of application atmospheric discharge in soil remediation. Most them focused on how improve degradation efficiency specific pollutants. In contrast, generation and development is less studied, which restricts widespread Pulse voltage widely used trigger soil, but requirement for power supply relatively high. The current paper proposes a new form voltage, with positive pulse superimposed DC bias. results show that existence...

Active magnetic bearing (AMB) system is a typical electromechanical which requires controller for stable levitation. disturbance rejection (ADRC) considered as an effective way to stabilize under external disturbance. And classical proportion-integration-differentiation (PID) still simple yet method most engineering applications its robustness. In this paper, ADRC and PID are implemented the levitation of AMB system. By mathematical analysis, connecting relationships control parameters...

Real-time diagnostics has always been a key point of plasma research. Optical emission spectroscopy is the most commonly used method to obtain parameters through fitting curves. But process time-consuming, making it almost impossible get data in real-time. In this paper, we propose using image recognition for real-time diagnostics. The relationship between visual images and rotational vibrational temperatures was established. A convolutional neural network model recognition. After training,...

On the basis of optimal allocation distributed generation in conventional distribution network, and considering characteristics active system influence controllable loads, a two-layer configuration model is established. The upper layer used to solve sizing power sources for systems lower strategy loads each period. output uncertainty uncontrollable generators, such as wind photovoltaic generation, characterized by box uncertain set based on robust optimization theory.With this model, can be...

Intensive care units (ICUs) adopted contamination control & air-conditioning technology to insure indoor air quality, which could afford upper cleanliness and comfort. Air conditioning ventilation ways affected greatly micro environment in ICUs. The paper discussed reasonable projects of the ICU using numerical simulation method. variation rules quality parameters ICUs were analyzed provide technical director for optimal designs under different supplying returning, numbers vents, types vents...

In the last decades, as a typical nonlinear system, active magnetic bearings (AMB) system has been widely applied in manufacturing systems. A sliding mode control (SMC) scheme for AMB is proposed with disturbance observation of linear extended state observer (LESO) this paper. The chattering reduced by LESO-SMC at least 60%. Sufficient BIBO (bounded input-bounded output) stability conditions closed-loop governed are derived Lyapunov theory. Finally, experiments conducted to verify...

With the development of human society, economy and technology, today's automobile industry is in an era rapidly changing, application autonomous driving-related technologies are full swing. This paper mainly focused on driving --- comparison between lidar computer vision cameras. Lidar cameras completely two different kinds solutions for auto-drive from all aspects. These schemes main current market, this essay focuses these by researching summarizing existing research online. Through...

This study proposes a novel method that integrates variational pattern decomposition (VMD), long short-term memory (LSTM) models, and particle swarm optimization (PSO) to optimize daily trading strategies for portfolio. By analyzing historical price data gold bitcoin, we have successfully identified the optimal investment actions (buy, hold, or sell) portfolio comprising different assets. The VMD-LSTM model proposed by us mitigates long-term dependence problem of machine learning algorithm...

As a novel structure, the integrated building-bridge station has appeared in high speed railway construction of China, which is great significant to ensure seismic safety new structure an earthquake. Meanwhile, isolation technique important passive control strategy also been widely used both buildings and bridges. Based on these, intensive study story carried out this paper investigate technique's feasibility effect station. Firstly, given special characteristics devices are usually...

Since there occur changes in temperature, density and pressure of sea water with increase depth difference areas, the buoyance submersible will change at different submergence depth. In order to implement coplex underwater tasks (for example search, salvage or rescure) deep submergence, small vehicles must be able submerse quickly safely into maintain good balance timely. It keep very slow zero cruising speed make depth-changing without overshoot. Microcomputer- based/integrated control...