Intelligent POIs Recommender System Based on Time Series Analysis with Seasonal Adjustment

Recommender systems have been applied on a variety of applications including movies, music, news, books, research articles, search queries, and travel information. Instead of searching travel information from the extremely huge amount of travel data, a personalized travel recommender system is desired. However, an inappropriate travel recommendation may result from a wrong season, even if it is already a correct location. The current recommender systems from time to time make an inappropriate commendation without considering the seasonal factor. In order to resolve the discrepancy, the seasonal factor should have been taken into consideration when making a good travel recommender system. Therefore, this study has taken the trend analysis, time series, and seasonal factor into considerations to cope with the above mentioned discrepancy and to make the travel recommender system renders a better fit.

MODELING AND FORECASTING DEMAND FOR A DIGITAL PRODUCT

Different approaches to modeling and forecasting the demand for a digital product, namely the paid activities of Facebook, are explored in the article. The company is given to reject its arrivals in the main form of advertisements. The available range of data on the company's activities allowed to build forecast models based on adaptive short-term forecasting methods, namely the Brown method and the adaptive multiplicative Holt-Winters model taking into account the quarterly seasonal factor. These models have the ability to continuously take into account the evolution of the dynamic characteristics of the studied processes, to adapt to these dynamics, giving weight and high information value to the available observations, if they are close to the current time. The models were tested for adequacy using a number of criteria, including the RS-test, the series criterion based on the median of the sample, the Student's t-test and the Darbin-Watson test. The comparative analysis of the obtained results by models allowed to choose a model that gives a fairly accurate result. The analysis also showed that there is a quarterly seasonality and, accordingly, a significant decline at the beginning of the year and income growth in recent quarters. The forecast for the 4th quarter of the next period is built. The forecast is compared with real data and the prospects for the development of digital products in Ukraine are determined, in particular, the spread of the use of digital services and products in many areas.

ESTIMATION OF THE VALUE OF THE COVID-19 BASIC REPRODUCTIVE NUMBER AND THE EFFECT OF ANTI-EPIDEMIC MEASURES AND “SEASONAL FACTOR” ON THIS VALUE

Background: The emergence of COVID-19 has led to increased attention to mathematical models of epidemiology, one of the main parameters of which is the basic reproductive number R0. Its value determines both the dynamics of the incidence and the level of anti-epidemic measures. Therefore, it is desirable to have a fairly accurate method for assessing R0 for each day. Aim: Develop a methodology for determining the R0 value for COVID-19, taking into account a reasonably rapid and significant change in its value over time. Methods: a method for calculating reproduction number R0 was proposed for assessing COVID-19 R0, taking into account the change in the contagiousness of the infected people during the infectious process. Results and Discussion: In Russia in June-August 2020, the reproduction rate was slightly less 1, but in September R0 began to grow, exceeded 1, which was caused a noticeable increase in the incidence. During June in Brazil, R0 stabilized at a value of 1. The activity of transmission of the pathogen is influenced by seasonal changes in the reproduction rate R0 and the level of community immune status. Based on the assessment of the dynamics of the incidence of other pneumonia, it was found that the change of R0 during the year is about 10%, with a minimum in summer. Conclusions: while maintaining the current activity of anti-epidemic measures due to seasonal factors of the activity of the transmission mechanism and accumulation of the immune individuals in the coming months, the situation in the Russian Federation will worsen with an increase in the incidence doubled within six months, and in Brazil in six months — improve with a decrease in the incidence ten times. However, the dynamics of the incidence will be determined primarily by the work of the administration and health authorities and the conscientiousness of citizens.

Monthly pork price forecasting method based on Census X12-GM(1,1) combination model

Background In recent years, the price of pork in China continues to fluctuate at a high level. The forecast of pork price becomes more important. Single prediction models are often used for this work, but they are not accurate enough. This paper proposes a new method based on Census X12-GM(1,1) combination model. Methods Monthly pork price data from January 2014 to December 2020 were obtained from the State Statistics Bureau(Mainland China). Census X12 model was adopted to get the long-term trend factor, business cycle change factor and seasonal factor of pork price data before September 2020. GM (1,1) model was used to fit and predict the long-term trend factor and business cycle change factor. The fitting and forecasting values of GM(1,1) were multiplied by the seasonal factor and empirical seasonal factor individually to obtain the fitting and forecasting values of the original monthly pork price series. Results The expression of GM(1,1) model for fitting and forecasting long-term trend factor and and business cycle change factor was X(1)(k) = −1704.80e−0.022(k−1) + 1742.36. Empirical seasonal factor of predicted values was 1.002 Using Census X12-GM(1,1) method, the final forecast values of pork price from July 2020 to December 2020 were 34.75, 33.98, 33.23, 32.50, 31.78 and 31.08 respectively. Compared with ARIMA, GM(1,1) and Holt-Winters models, Root mean square error (RMSE), mean absolute percentage error (MAPE) and mean absolute error (MAE) of Census X12-GM(1,1) method was the lowest on forecasting part. Conclusions Compared with other single model, Census X12-GM(1,1) method has better prediction accuracy for monthly pork price series. The monthly pork price predicted by Census X12-GM(1,1) method can be used as an important reference for stakeholders.

Ifluence of seasonity on vitamin D level in patients with HCV infection and healthy people

According to WHO, about 150-200 million people are currently infected with the HCV virus worldwide. Recently, in the professional literature, the number of publications on the role of vitamin D in patients with viral hepatitis C has increased as vitamin D metabolism occurs with the participation of the liver and its deficiency is associated with an increased risk of infectious diseases. The aim of this study was to investigate the effect of seasonal factor on vitamin D (25 hydroxycalciferol) levels in patients with chronic viral hepatitis C and healthy subjects. The study involved 100 patients in the registry of patients with chronic viral hepatitis in the Dnipropetrovsk region. The prevalence and deficiency of vitamin D in patients with chronic viral hepatitis C and conditionally healthy subjects at different times of the year were determined, which showed the presence or absence of a seasonal effect on serum 25(OH)D level. Patients were divided into two groups, depending on the time of the year (autumn-winter and spring-summer), in which the level of 25 (OH) D was determined. The serum was metabolised by vitamin D, which is synthesized by the liver – 25 hydroxycalciferol (25 (OH) D), an indicator of the supply of vitamin D to the human body. Vitamin D levels were evaluated according to the M.F. Holick classification. According to the level of vitamin D patients were divided into 3 groups (patients with normal level, insufficient (suboptimal) level and vitamin D deficiency). The results of the study showed no effect of seasonal factor on the level of 25 (OH) D in the serum of patients with chronic viral hepatitis C. Vitamin D levels are controlled by the time of the year: in spring and summer this indicator is normal, in autumn and winter – seasonal decrease in vitamin D.

A PSO algorithm-based seasonal nonlinear grey Bernoulli model with fractional order accumulation for forecasting quarterly hydropower generation

The hydropower plays a key role in electricity system owing to its renewability and largest share of clean electricity generation that promotes sustainable development of national economy. Developing a proper forecasting model for the quarterly hydropower generation is crucial for associated energy sectors, which could assist policymakers in adjusting corresponding schemes for facing with sustained demands. For this purpose, this paper presents a fractional nonlinear grey Bernoulli model (abbreviated as FANGBM(1,1)) coupled seasonal factor and Particular Swarm Optimization (PSO) algorithm, namely PSO algorithm-based FASNGBM(1,1) model. In the proposed method, the moving average method that eliminates the seasonal fluctuations is introduced into FANGBM(1,1), then in which the structure parameters of FASNGBM(1,1) are determined by PSO. Based on hydropower generation of China from the first quarter of 2011 to the final quarter of 2018 (2011Q1-2018Q4), the numerical results show that the proposed model has a better performance than that of other benchmark models. Eventually, the quarterly hydropower generation of China from 2019 to 2020 are forecasted by the proposed model, according to results, the hydropower generation of China will reach 11287.14 × 108 Kwh in 2020.

The Seasonal Factor in Annual Milk Revenue

The article examines the influence of the seasonality factor on changes in the price and volume of milk in Russia. The contribution of changes in prices and volumes to the seasonality of proceeds from the sale of raw milk are analyzed. The reasons for the seasonal decline in milk production and ways of reducing the influence of the seasonality factor in dairy production are discussed. Based on statistics from Rosstat and analytical agencies, the presence of seasonality in dairy production, in which the volume of milk production plays the main role, is revealed.

Feed microbial contamination in pig-breeding: Modern threats and ways to overcome them

The paper presents the results on the species and percentage composition of the microflora in biofilms of pig feed, which varies depending on the seasonal factor. Bacteria Streptococcus spp., Pasteurella multocida, Neisseria spp., and Clostridium perfringens in biofilms were found much more often (by 25% or more) in the warm period of the year, while listeria in silage and haylage — in the autumn–winter period. This property of feed biofilms is also significantly influenced by the conditions of cultivation, harvesting and storage of agricultural products. In the study of biofilms of microflora of barley, corn and wheat, it was found that their structural basis are aerobic fungi of the mold Aspergillus spp. Bacteria Streptococcus spp., Pasteurella multocida, Neisseria spp., and Clostridium perfringens without mold form much looser biofilms in vitro and these biofilms are much more sensitive to a wide range of commercial antibiotics. The structural basis of polymicrobial biofilms of barley, corn and wheat microflora is highly likely to be aerobic fungi of Aspergillus spp.