The proposed method outperforms others in quantity classification, achieving the accuracies of 88%, 90%, and 84% for the random woodland, SVM, and K-NN correspondingly. To overcome see more these problems, a substation gear temperature forecast strategy is proposed centered on multivariate information fusion, convolutional neural system (CNN) and gated recurrent unite (GRU) in this article Sensors and biosensors . Firstly, in line with the correlation analysis including linear correlation mapping, autocorrelation function and limited autocorrelation function for substation gear temperature information, the function vectors from background, time and area are determined, that’s the multivariate information fusion function vector (denoted as MIFFV); next, the dimension of MIFFV is paid down by principal component analysis (PCA), extract some of the most important functions and form the reduced feature vector (denoted as RFV); then, CNN is used for deep understanding how to draw out the connection bTaizhou City, Zhejiang Province is conducted because of the technique suggested in this specific article. Through the comparative research through the two facets of functions and methods, under the two forecast performance analysis indexes of mean absolute percentage error (MAPE) and root-mean-square mistake (RSME), two primary conclusions are attracted (1) MIFFV from three components of background functions, time features infected false aneurysm and space features have much better prediction performance compared to the solitary function vector and also the combined feature vector of two aspects; (2) compared with other four related models beneath the exact same conditions, RFV is regarded as the input associated with designs, the suggested model has better forecast performance.Curcuma longa (turmeric) and Curcuma zanthorrhiza (temulawak) tend to be members of the Zingiberaceae family that contain curcuminoids, essential oils, starch, protein, fat, cellulose, and minerals. The nutritional content proportion of turmeric is significantly diffent from temulawak which indicates variations in financial worth. However, only a few individuals who realize organic plants, can recognize the difference between them. This research is designed to develop a model that will differentiate amongst the two types of Zingiberaceae in line with the picture grabbed from a mobile phone camera. An accumulation of pictures composed of both forms of rhizomes are used to develop a model through a learning process utilizing transfer understanding, specifically pre-trained VGG-19 and Inception V3 with ImageNet weight. Experimental outcomes reveal that the precision prices associated with the designs to classify the rhizomes tend to be 92.43% and 94.29%, consecutively. These accomplishments are quite encouraging to be utilized in various useful use.During unprecedented activities such as COVID-19, the material of culture comes under anxiety and all stakeholders need increase the predictability of the future and minimize the ongoing uncertainties. In this analysis, an endeavor was meant to model the situation where the belief “trust” is calculated so as to map the behavior of society. Nonetheless, officially, the purpose of this scientific studies are to not determine the “degree of rely upon community” as a consequence of some certain thoughts or sentiments that the community is experiencing at any certain time. This task can be involved because of the construction of a computational model that can help in improving our comprehension of the characteristics of digital communities, especially when it comes to the mindset called “trust.” The digital society trust analysis (D.S.T.A.) design that is supplied is simple to configure and easy to make usage of. It offers many past designs, such as standing models, Schelling’s type of segregation, and tipping points, in order to build models for understanding the dynamics of a society reeling underneath the outcomes of a COVID-19 pandemic, misinformation, phony development, along with other sentiments that impact the behavior regarding the different groups.As among the significant platforms of interaction, social networks have become a valuable source of opinions and thoughts. Given that sharing of emotions offline and on the internet is rather comparable, historic posts from social networking sites be seemingly a valuable source of data for calculating observable subjective wellbeing (OSWB). In this study, we calculated OSWB indices when it comes to Russian-speaking segment of Twitter utilising the Affective Social information Model for Socio-Technical Interactions. This design utilises demographic information and post-stratification ways to make the information sample agent, by selected traits, associated with the basic population of a country. For sentiment evaluation, we fine-tuned RuRoBERTa-Large on RuSentiTweet and achieved brand new state-of-the-art results of F1 = 0.7229. Several calculated OSWB indicators demonstrated reasonable Spearman’s correlation with all the standard survey-based net affect (rs = 0.469 and rs = 0.5332, p less then 0.05) and positive influence (rs = 0.5177 and rs = 0.548, p less then 0.05) indices in Russia.To solve the nonlinear constrained optimization issue, a particle swarm optimization algorithm based on the improved Deb criterion (CPSO) is proposed.