Neural network forecasting of news feeds

Osipov, Vasiliy; Kuleshov, S.; Zaytseva, Alexandra A.; Levonevskiy, Dmitriy; Miloserdov, Dmitriy

doi:10.1016/j.eswa.2020.114521

Cited by 11 publications

(3 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The neurons of this layer serve as independent variables of any physical system. A separate class of combined RNNs should systems built on the basis of using two identical recurrent neural networks with controlled elements (Osipov and Osipova 2018) and a control unit for these networks (Osipov et al 2020(Osipov et al , 2021(Osipov et al , 2022. A generalized structure of such combined RNNs is shown in Fig.…”

Section: Esquivel Et Al (2020) Proposes a Convolutionalmentioning

confidence: 99%

Data fusion using neural networks for intelligent multilevel control of artificial and natural objects: a survey

Tianxing

Osipov

Zhukova

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

In the modern world, data fusion (DF) technique has been playing an increasingly important role for effective and efficient data transformation in subject domains. DF aims to solve the problem of control of artificial and natural objects. Although DF has been intensively studied, neither unified problem statements of the fusion of the heterogeneous, multimodal, complex, imperfect data nor in-depth elaboration of characteristics and challenges about the application of DF on such specific situations are available. It is also unclear to what extent DF challenges have been addressed and what the overall research landscape of DF is. We define the requirements for DF in the interests of intelligent control of artificial and natural objects and look at the applications, reveal the capabilities of traditional DF methods and analyze the advantages that neural networks (NN) have over them. We show that despite of the evident advantages, none of the existing NNs satisfy all the requirements of DF. Based on the analysis of NN architectures, we determine the requirements for advanced NN architectures and discuss future research directions.

show abstract

Section: Esquivel Et Al (2020) Proposes a Convolutionalmentioning

confidence: 99%

Data fusion using neural networks for intelligent multilevel control of artificial and natural objects: a survey

Tianxing

Osipov

Zhukova

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Attempts have been made to assess the impact of information sources on people's situational awareness and social distancing (Wu et al, 2012;Qazi et al, 2020;Tiwari et al, 2021), including in the context of media coverage of the COVID-19 pandemic and early proliferation of the infection in China (Liu et al, 2020;Zhou et al, 2020). A number of models exist, each with its own advantages and disadvantages under particular conditions (Chubb, Jacobsen, 2009;Nadella et al, 2020), which help predict the development of epidemics (Newbold, Granger, 1974;Holko et al, 2016;Holko et al, 2020;Hu et al, 2020;Levashkin et al, 2021;Medrek, Pastuszak, 2021;Katris, 2021;Osipov et al, 2021). 1 Various approaches are applied to monitor the situation and process statistical data in order to set parameters and initial states for the application of these models, but little attention is paid to assessing the potential impact of sharing epidemiological information on public health.…”

Section: Introductionmentioning

confidence: 99%

Epidemiological Informing of the Population in Cities: Models and Their Application

Osipov¹,

Osipova²,

Kuleshov³

et al. 2022

Foresight and STI Governance

View full text Add to dashboard Cite

With an increase of population density and contacts between people, the emergence of new biological viruses, the threat of various epidemics is growing. Countering these threats involves the implementation of large-scale preventive, therapeutic and other measures, both before the start and during the epidemic. Epidemiological informing of the population plays an important role in such counteraction. The currently used models of epidemiological informing the population of cities largely do not meet the needs of practice. This negatively affects the effectiveness of the response to epidemics. The purpose of the study is to develop new models and justify their applicability for understanding the processes in public health, the impact of epidemics on the economy and business. For the quantitative substantiation of programs (scenarios), such epidemiological informing, a method based on new models of epidemic development in related cities is proposed. The method is characterized by a new objective function that links economic efficiency with the state of health of the population in an epidemic. The models differ from the known solutions both in the space of the selected states of the processes under study and in the connections between them.Using the developed method, seven possible programs of epidemiological informing the population of related cities were analyzed and the best of them was found for specific conditions. New regularities have been established between the parameters of the programs being implemented and the results of the impact on the health and performance capability of the population. It is shown how an epidemic can develop in cities that are differently connected to each other by vehicles. The proposed method allows quickly find the best epidemiological informing programs for the population. The models underlying this method make it possible to predict public health and the impact of epidemics on the economy and business, depending on the planned measures to counteract epidemics. They are also applicable to determine the sources and time of infections’ onset. The obtained simulation results are in good agreement with the known facts. The method can be applied in advanced information systems to support the adoption of far-sighted decisions to counteract epidemics.

show abstract

“…Prediction must be carried out using the appropriate methods [6]. The methods for time series prediction are based on the idea that historical data contain inherent patterns that provide useful information for the future description of the phenomenon under consideration [7].…”

Section: Introductionmentioning

confidence: 99%

Implementation of Particle Swarm Optimization (PSO) to Improve Neural Network Performance in Univariate Time Series Prediction

Tyas¹,

Setianama²,

Fajriyah³

et al. 2021

KINETIK

View full text Add to dashboard Cite

One of the oldest known predictive analytics techniques is time series prediction. The target in time series prediction is use historical data about a specific quantity to predicts value of the same quantity in the future. Multivariate time series (MTS) data has been widely used in time series prediction research because it is considered better than univariate time series (UTS) data. However, in reality MTS data sets contain various types of information which makes it difficult to extract information to predict the situation. Therefore, UTS data still has a chance to be developed because it is actually simpler than MTS data. UTS prediction treats forecasts as a single variable problem, whereas MTS may employ a large number of time-concurred series to make predictions. Neural Network (NN) model could be built to predict the target variable given the other (predictor) variables. In this study, we used Particle Swarm Optimization (PSO) algorithm to optimize performance of NN on a UTS dataset. Our proposed model is validated using x-validation and and use RMSE to measure its performance. The experimental results show that NN performance after optimization using PSO produces good results compared to classical NN performance. This is evidenced by the value of RMSE = 0.410 which is the smallest RMSE value produced. The smaller the RMSE value, the better the model performance. It can be concluded that the proposed method can improve NN performance on UTS data.

show abstract

Neural network forecasting of news feeds

Cited by 11 publications

References 26 publications

Data fusion using neural networks for intelligent multilevel control of artificial and natural objects: a survey

Data fusion using neural networks for intelligent multilevel control of artificial and natural objects: a survey

Epidemiological Informing of the Population in Cities: Models and Their Application

Implementation of Particle Swarm Optimization (PSO) to Improve Neural Network Performance in Univariate Time Series Prediction

Contact Info

Product

Resources

About