Author identification of literary works based on text analysis and deep learning.

With the development of science, speech, picture, and other analysis, problems have been gradually better solved, but the study of Chinese text has been a complex problem to overcome. Chinese text analysis requires not only statistics but also semantic comprehension analysis. Different text types need other language style feature modeling to obtain good recognition results. In this study, we use the deep learning method to construct an automatic text feature extraction model and classify it with the author as a classification label. This study presents a literature author recognition model based on deep learning, which is mainly divided into three phases: text preprocessing, feature extraction, and classification. Each part consists of several small modules or steps. First, we input the corpus to Word2Vec to generate the new word vector. Then, the improved text feature extractor based on CNN and Attention extracts the text features and uses them as the input of the CNN convolution layer. After convolution, the text is combined with bits to get Window Feature Sequence. It is the text feature vector. Next, based on LSTM and Softmax classification output, Window Feature Sequence is used as the input of LSTM to obtain two one-dimensional vectors spliced by concatenate layer. Finally, the result is classified through the fully connected layer, Batch Normalization layer, and Softmax. The performance of the proposed model in recognizing authors of Chinese literature was evaluated using two datasets. In the research process, the data we collected included works of different forms, such as prose and fiction. The research results show that the proposed model can effectively identify author identity. The classification accuracy of our proposed algorithm is significantly better than that of the benchmark model.

Parameter-Free State Estimation Based on Kalman Filter with Attention Learning for GPS Tracking in Autonomous Driving System.

GPS-based maneuvering target localization and tracking is a crucial aspect of autonomous driving and is widely used in navigation, transportation, autonomous vehicles, and other fields.The classical tracking approach employs a Kalman filter with precise system parameters to estimate the state. However, it is difficult to model their uncertainty because of the complex motion of maneuvering targets and the unknown sensor characteristics. Furthermore, GPS data often involve unknown color noise, making it challenging to obtain accurate system parameters, which can degrade the performance of the classical methods. To address these issues, we present a state estimation method based on the Kalman filter that does not require predefined parameters but instead uses attention learning. We use a transformer encoder with a long short-term memory (LSTM) network to extract dynamic characteristics, and estimate the system model parameters online using the expectation maximization (EM) algorithm, based on the output of the attention learning module. Finally, the Kalman filter computes the dynamic state estimates using the parameters of the learned system, dynamics, and measurement characteristics. Based on GPS simulation data and the Geolife Beijing vehicle GPS trajectory dataset, the experimental results demonstrated that our method outperformed classical and pure model-free network estimation approaches in estimation accuracy, providing an effective solution for practical maneuvering-target tracking applications.

Predicting the monthly consumption and production of natural gas in the USA by using a new hybrid forecasting model based on two-layer decomposition.

As an efficient, economical, and clean energy, natural gas plays an important role in the development of the new energy revolution. Accurate prediction of natural gas consumption and production can adjust energy deployment in advance, which can ensure the stable operation of natural gas. Considering the complex and non-linear characteristics of natural gas production and consumption data, this paper develops a new hybrid forecasting model (WPD-VMD-LSTM) based on the fuzzy entropy, variational mode decomposition (VMD), wavelet packet decomposition (WPD), and Long Short-Term Memory (LSTM). In this model, WPD and VMD undertake the tasks of primary and secondary decompositions, respectively; fuzzy entropy is used for the preprocessing process before the re-decomposition; and LSTM is used to predict the decomposed time series. In particular, the different criteria set by fuzzy entropy lead to the establishment of two prediction models. Then, two models are used to study monthly natural gas consumption and production in the USA. The results demonstrate that the proposed model performs significantly better than other comparable models and the target model has some practical value. Meanwhile, models may cope with different types of energy data, and models can accurately predict energy transformations with strong applicability, which can be applied to future energy forecasting in various fields. Finally, the constructed models are used to forecast the NGC and NGP in the USA in the next 3 years and make reasonable policy recommendations based on the forecast results.

[Simulation Analysis of the Impact of COVID-19 Pandemic Control on Air Quality in Henan Province based on Machine Learning Algorithm].

To reduce the risks of COVID-19 on society and the health of the general public, necessary prevention and control measures were implemented throughout China in 2020. Consequently, air quality was greatly improved due to lower emissions. However, the improvement of air quality could also be closely related to meteorological conditions. During quarantine (January 27 to February, 2020), reductions were observed in the concentration of all air pollutants in Henan Province (PM2.5, PM10, SO2, CO, and NO2 decreased by 36.89%, 34.18%, 19.43%, 29.85%, and 58.51%, respectively) relative to measurements taken from January 1 to 26, 2020. The only exception was for the concentration of O3, which increased by 69.64%. This study evaluates the importance of meteorological conditions in air pollution, through simulation with a long-and-short-term memory network (LSTM) and a machine learning algorithm. Results show that meteorological conditions play a crucial role in air pollutant formation. Given favorable meteorological factors, the concentrations of pollutants could be reduced by 15%-30%, while the reduction due to anthropogenic emission control ranges from 6%-40%. During the epidemic, meteorological conditions and human emissions accounted for 34.84% and 34.81% of the increase in O3 concentration, respectively. The results show that primary pollutant concentrations are more sensitive to the intensity of anthropogenic emissions. However, secondary pollutants are more dependent on meteorological factors. Furthermore, a nonlinear relationship has been identified between O3 concentration and to emission intensity. Further investigation into the causes of the rise in O3 concentration is necessary to gain a greater understanding and better control of particulate matter and O3 pollution.