Search results

Social media provide an impressive amount of data about users and their interactions, thereby offering computer and social scientists, economists, and statisticians – among others – new opportunities for research. Arguably, one of the most interesting lines of work is that of predicting future events and developments from social media data. However, current work is fragmented and lacks of widely accepted evaluation approaches. Moreover, since the first techniques emerged rather recently, little is known about their overall potential, limitations and general applicability to different domains. Therefore, better understanding the predictive power and limitations of social media is of utmost importance.

Different types of forecasting models and their adaptation to the special circumstances of social media are analyzed and the most representative research conducted up to date is surveyed. Presentations of current research on techniques, methods, and empirical studies aimed at the prediction of future or current events from social media data are provided.

A taxonomy of prediction models is introduced, along with their relative advantages and the particular scenarios where they have been applied to. The main areas of prediction that have attracted research so far are described, and the main contributions made by the papers in this special issue are summarized. Finally, it is argued that statistical models seem to be the most fruitful approach to apply to make predictions from social media data.

This special issue raises important questions to be addressed in the field of social media-based prediction and forecasting, fills some gaps in current research, and outlines future lines of work.

Location‐prediction enables the next generation of location‐based applications. The purpose of this paper is to provide a historical summary of research in personal location‐prediction. Location‐prediction began as a tool for network management, predicting the load on particular cellular towers or WiFi access points. With the increasing popularity of mobile devices, location‐prediction turned personal, predicting individuals' next locations given their current locations.

This paper includes an overview of prediction techniques and reviews several location‐prediction projects comparing the raw location data, feature extraction, choice of prediction algorithms and their results.

A new trend has emerged, that of employing additional context to improve or expand predictions. Incorporating temporal information enables location‐predictions farther out into the future. Appending place types or place names can improve predictions or develop prediction applications that could be used in any locale. Finally, the authors explore research into diverse types of context, such as people's personal contacts or health activities.

This overview provides a broad background for future research in prediction.

Since crude oil is crucial to the nation's economic growth, crude oil futures are closely related to many other markets. Accurate forecasting can offer investors trustworthy guidance. Numerous studies have begun to consider creating new metrics from social networks to improve forecasting models in light of their rapid development. To improve the forecasting of crude oil futures, the authors suggest an integrated model that combines investor sentiment and attention.

This study first creates investor attention variables using Baidu search indices and investor sentiment variables for medium sulfur crude oil (SC) futures by collecting comments from financial forums. The authors feed the price series into the NeuralProphet model to generate a new feature set using the output subsequences and predicted values. Next, the authors use the CatBoost model to extract additional features from the new feature set and perform multi-step predictions. Finally, the authors explain the model using Shapley additive explanations (SHAP) values and examine the direction and magnitude of each variable's influence.

The authors conduct forecasting experiments for SC futures one, two and three days in advance to evaluate the effectiveness of the proposed model. The empirical results show that the model is a reliable and effective tool for predicting, and including investor sentiment and attention variables in the model enhances its predictive power.

The data analyzed in this paper span from 2018 through 2022, and the forecast objectives only apply to futures prices for those years. If the authors alter the sample data, the experimental process must be repeated, and the outcomes will differ. Additionally, because crude oil has financial characteristics, its price is influenced by various external circumstances, including global epidemics and adjustments in political and economic policies. Future studies could consider these factors in models to forecast crude oil futures price volatility.

In conclusion, the proposed integrated model provides effective multistep forecasts for SC futures, and the findings will offer crucial practical guidance for policymakers and investors. This study also considers other relevant markets, such as stocks and exchange rates, to increase the forecast precision of the model. Furthermore, the model proposed in this paper, which combines investor factors, confirms the predictive ability of investor sentiment. Regulators can utilize these findings to improve their ability to predict market risks based on changes in investor sentiment. Future research can improve predictive effectiveness by considering the inclusion of macro events and further model optimization. Additionally, this model can be adapted to forecast other financial markets, such as stock markets and other futures products.

The authors propose a novel integrated model that considers investor factors to enhance the accuracy of crude oil futures forecasting. This method can also be applied to other financial markets to improve their forecasting efficiency.

This study predicts the growth of Africa's international trade from 2011 to 2040 by accounting for the uncertainties in the continent.

This study applies a scenario planning method (SPM) to develop multiple future scenarios considering uncertainties inherent in African socio-economies related to the success or failure of economic and industrial policies (EIPs) and economic corridor development policies (ECDPs). Subsequently, based on these future scenarios, the growth of African international trade from 2011 to 2040 is predicted using the Global Trade Analysis Project (GTAP) model.

The predictions reveal that if the EIPs and the ECDPs are successfully implemented, Africa, as a whole, will experience a significant increase in trade, estimated at US$ 1,905 billion and US$ 1,599 billion for exports and imports, respectively, compared to the scenario in which they fail. However, the effects vary greatly by country or region and industrial sector. The results also show that African intra-regional trade is rapidly expanding and is the second-largest after trade with Europe followed by other continents.

SPM, which allows us to reflect the uncertainties affecting African international trade prediction, is applied to build the future scenarios. The study comprehensively predicts African future international trade by setting a wide range of exogenous variables and parameters (input conditions for the GTAP model) related to EIPs and ECDPs.

This paper focuses on assembly line performance of an automotive body shop that builds body‐in‐white (BIW) assembly utilizing about 700+ process robots. These robots perform various operations such as welding, sealing, part handling, stud welding and inspection. There is no accurate tool available for the plant personnel to predict the future throughput based on plant's data. The purpose of this paper is to provide future throughput performance prediction based on plant data using Box‐Jenkins' ARMA model.

The following data were collected for five major assembly lines. First, the assembly machine‐in‐cycle time: the assembly line machines include robots that perform various functions like load, welding or sealing and unloading parts; the manual operators loading cycle time to the production fixtures. The conveyors act as buffers in between stations, and also feed to the production cells, and carry parts from station to station. The conveyors' downtime and uptime were also part of the machine‐in‐cycle time; second, the number of units produced from the beginning to the end of the assembly line; third, the number of fault occurrences in the assembly line due to various machine breakdowns; fourth, the machine availability percentage – i.e. the machine is readily available to perform its functions (the machine blocked upstream (starving) and blocked down (downstream) state is considered here); fifth, the actual efficiency of the machine measured in percentage based on output percentage; sixth, the expected number of units at designed efficiency.

In summary, this research paper provided a systematic development of a forecast model based on Box‐Jenkin's ARMA methodology to analyze the complex assembly line process performance data. The developed ARMA forecast models proved that the future prediction can be accurately predicted based on the past plant performance data. The developed ARMA forecast models predicted the future throughput performance within 99.52 percent accuracy. The research findings were validated by the actual plant performance data.

In this study, the automotive assembly process machines (robots, conveyors and fixtures) production data were collected, statistically analyzed and verified for viable ARMA model verification. The verified ARMA model has been used to predict the plant future months' throughput with 99.52 percent accuracy, based on the plant production data. This research is unique because of its practical usage to improve production.

The purpose of this study is to provide a novel portfolio asset prediction by means of the modified deep learning and hybrid meta-heuristic concept. In the past few years, portfolio optimization has appeared as a demanding and fascinating multi-objective problem, in the area of computational finance. Yet, it is accepting the growing attention of fund management companies, researchers and individual investors. The primary issues in portfolio selection are the choice of a subset of assets and its related optimal weights of every chosen asset. The composition of every asset is chosen in a manner such that the total profit or return of the portfolio is improved thereby reducing the risk at the same time.

This paper provides a novel portfolio asset prediction using the modified deep learning concept. For implementing this framework, a set of data involving the portfolio details of different companies for certain duration is selected. The proposed model involves two main phases. One is to predict the future state or profit of every company, and the other is to select the company which is giving maximum profit in the future. In the first phase, a deep learning model called recurrent neural network (RNN) is used for predicting the future condition of the entire companies taken in the data set and thus creates the data library. Once the forecasting of the data is done, the selection of companies for the portfolio is done using a hybrid optimization algorithm by integrating Jaya algorithm (JA) and spotted hyena optimization (SHO) termed as Jaya-based spotted hyena optimization (J-SHO). This optimization model tries to get the optimal solution including which company has to be selected, and optimized RNN helps to predict the future return while using those companies. The main objective model of the J-SHO-based RNN is to maximize the prediction accuracy and J-SHO-based portfolio asset selection is to maximize the profit. Extensive experiments on the benchmark datasets from real-world stock markets with diverse assets in various time periods shows that the developed model outperforms other state-of-the-art strategies proving its efficiency in portfolio optimization.

From the analysis, the profit analysis of proposed J-SHO for predicting after 7 days in next month was 46.15% better than particle swarm optimization (PSO), 18.75% better than grey wolf optimization (GWO), 35.71% better than whale optimization algorithm (WOA), 5.56% superior to JA and 35.71% superior to SHO. Therefore, it can be certified that the proposed J-SHO was effective in providing intelligent portfolio asset selection and prediction when compared with the conventional methods.

This paper presents a technique for providing a novel portfolio asset prediction using J-SHO algorithm. This is the first work uses J-SHO-based optimization for providing a novel portfolio asset prediction using the modified deep learning concept.

This paper aims to explore the determinants of citizens’ future use of mobile applications provided by government. Research on citizen-initiated contacts with government has focused on both non-technology and technology related contacts. Existing research, however, has not examined the impact of mobile applications or “apps” on citizen-initiated contacts with government. Furthermore, existing research has not examined satisfaction with mobile government and whether this impacts future use.

The authors examine future use of mobile apps through an empirical analysis of a public opinion survey of citizen users in four of the largest cities in China (Beijing, Shanghai, Guangzhou and Shenzhen).

Using ordered logistic regression analysis, this study found that the strongest predictors of future use were demand and satisfaction with mobile apps. However, there was no wide-scale evidence of socioeconomic status and age impacting mobile apps future use.

The findings in this study contribute to both theory and practice of the determinants of mobile government adoption.

The results challenge the citizen-initiated contact theory, as socioeconomic status was not a major predictor of mobile apps future use in China. The results further indicate that satisfaction was a good predictor of mobile apps future use.

The purpose of the article is to contribute to the body of research on digital transformation among members of the supply chain operating in an emerging economy. This paper researches the digital transformation trends happening across Vietnamese supply chains, by investigating the current adoption rates, predicted impact levels and financial investments being made in key Industry 4.0 technologies.

By using a semi-structured online survey, the experiences of 281 supply chain professionals in Vietnam were captured. Subsequently, statistical techniques examining variances in means, regression analysis and Monte Carlo simulation were applied.

The findings of this study offer a comprehensive understanding of Industry 4.0 technology in Vietnam, highlighting the prevalent technologies being prioritized. Big data analytics and the Internet of things are expected to have the most substantial impact on businesses over the next 5–10 years and have received the most financial investment. Conversely, Blockchain is perceived as having less potential for future investment. The study further identifies several technological synergies, such as combining advanced robotics, artificial intelligence and the Internet of things to build effective and flexible factories, that can lead to more comprehensive solutions. It also extends diffusion of innovation theory, encompassing investment and impact considerations.

This study offers valuable insights into the impact and financial investment in Industry 4.0 technologies by Vietnamese supply chain firms. It provides a theoretical contribution via an extension of the diffusion of innovation theory and contributes toward a better understanding of the current Industry 4.0 landscape in developing economies. The findings have significant implications for future managerial decision-making, on the impact, viability and resourcing needs when undertaking digital transformation.

This paper aims to analyze the socioeconomic impacts of infectious diseases based on uncertain behaviors of social and effective subsystems in the countries. The economic impacts of infectious diseases in comparison with predicted gross domestic product (GDP) in future years could be beneficial for this aim along with predicted social impacts of infectious diseases in countries.

The proposed uncertain SEIAR (susceptible, exposed, infectious, asymptomatic and removed) model evaluates the impacts of variables on different trends using scenario base analysis. This model considers different subsystems including healthcare systems, transportation, contacts and capacities of food and pharmaceutical networks for sensitivity analysis. Besides, an adaptive neuro-fuzzy inference system (ANFIS) is designed to predict the GDP of countries and determine the economic impacts of infectious diseases. These proposed models can predict the future socioeconomic trends of infectious diseases in each country based on the available information to guide the decisions of government planners and policymakers.

The proposed uncertain SEIAR model predicts social impacts according to uncertain parameters and different coefficients appropriate to the scenarios. It analyzes the sensitivity and the effects of various parameters. A case study is designed in this paper about COVID-19 in a country. Its results show that the effect of transportation on COVID-19 is most sensitive and the contacts have a significant effect on infection. Besides, the future annual costs of COVID-19 are evaluated in different situations. Private transportation, contact behaviors and public transportation have significant impacts on infection, especially in the determined case study, due to its circumstance. Therefore, it is necessary to consider changes in society using flexible behaviors and laws based on the latest status in facing the COVID-19 epidemic.

The proposed methods can be applied to conduct infectious diseases impacts analysis.

In this paper, a proposed uncertain SEIAR system dynamics model, related sensitivity analysis and ANFIS model are utilized to support different programs regarding policymaking and economic issues to face infectious diseases. The results could support the analysis of sensitivities, policies and economic activities.

• A new system dynamics model is proposed in this paper based on an uncertain SEIAR model (Susceptible, Exposed, Infectious, Asymptomatic, and Removed) to model population behaviors;

• Different subsystems including healthcare systems, transportation, contacts, and capacities of food and pharmaceutical networks are defined in the proposed system dynamics model to find related sensitivities;

• Different scenarios are analyzed using the proposed system dynamics model to predict the effects of policies and related costs. The results guide lawmakers and governments' actions for future years;

• An adaptive neuro-fuzzy inference system (ANFIS) is designed to estimate the gross domestic product (GDP) in future years and analyze effects of COVID-19 based on them;

• A real case study is considered to evaluate the performances of the proposed models.

A new system dynamics model is proposed in this paper based on an uncertain SEIAR model (Susceptible, Exposed, Infectious, Asymptomatic, and Removed) to model population behaviors;

Different subsystems including healthcare systems, transportation, contacts, and capacities of food and pharmaceutical networks are defined in the proposed system dynamics model to find related sensitivities;

Different scenarios are analyzed using the proposed system dynamics model to predict the effects of policies and related costs. The results guide lawmakers and governments' actions for future years;

An adaptive neuro-fuzzy inference system (ANFIS) is designed to estimate the gross domestic product (GDP) in future years and analyze effects of COVID-19 based on them;

A real case study is considered to evaluate the performances of the proposed models.

The purpose of this paper is to examine empirical customer account data from 2006 through 2012 to review the probability of default (PD) rating methodology implemented by a FCS association for production agricultural accounts. This analysis provides insight into the migration of accounts across the association’s currently established PD rating categories with negative migration being a precursor to potential loan default.

The data set contained 17,943 observations from the years 2006 to 2012 and consisted of various fields of data including balance sheet date, earnings statement date, and PD rating as of the statement date. The methods include analysis on the dynamics of the PD ratings and component ratios. OLS regression was used to analyze the data to see how the current period PD rating and component ratios affected the PD rating one year, three years, and five years out. OLS regression examined the statistical significance of the PD ratings and ratio components for this analysis. The dependent variable, Future PD Rating, represents the assigned PD rating for the observed farm either one, three, or five years into the future. It is expected that the initial PD rating in any given year would have a positive relationship, and be statistically significant in estimating future PD ratings. The independent variables are the current PD rating and the various component ratios of the inverse current ratio (CR), the debt to asset ratio (D/A), the gross profit to total liabilities ratio, the inverse debt coverage ratio, working capital to gross profit, and funded debt to EBITDA.

Results indicate that financial ratio information gathered today can do a good job forecasting PD ratings up to three years in the future. CR information does not forecast five years into the future very well. Thus, there is an important need to update financial information on a regular basis. The results indicate that the D/A information is very important in predicting risk ratings. As the production agriculture sector has experienced difficult financial conditions during 2014 and 2015, agricultural finance institutions need to obtain up-to-date financial information from their clientele to effectively assess the risk of and manage their financial portfolio.

Several previous works have examined and established models to assess risk in agricultural lending. This research adds to this body of work by examining the migration of an account’s risk-rating class over time using actual lender account data.