Search

Digitalization affects all fields of the modern world including the economic, political, and social aspects of our life. Governments are also involved in this process. Considering the rapid transition towards digital e-services that we evidence since the COVID-19 outbreak, the detection and analysis of public opinion on e-government services becomes even more important. The sentiments and opinions of users on digital public services might be used to foster improvements in the development and implementation of the e-services. In this paper we engage the advances in NLP so as to examine the possibilities to analyze in an automated manner the opinions and sentiments towards e-government services expressed by citizens in various social networks in Bulgaria and this is the main goal of our research. For this purpose, we design an integrated ML-based AI system that aims to support decision makers in e-government and public services provision. The system utilizes a variety of data sources - news websites, web forums and other online social networks. To the best of the authors' knowledge, this is the first study that develops a methodology for mining public opinion in Bulgaria based on a combination of NLP and ML techniques, rather than relying on surveys or in-depth interviews. [ABSTRACT FROM AUTHOR]

The United Nations recently published the "E-government survey 2020" with the main aim of assessing the e-government development status of all United Nations member states. The survey outlines 14 leading countries in e-government development (out of 193 member states) some of them claiming to utilize technologies as artificial intelligence (AI), big data and blockchain. Moreover, with the burst of the COVID-19 pandemic the topic on development and implementation of e-government services becomes even hotter. However, along with the research on the process of digitalization of public services, it is important to develop tools measuring how these rapid changes are perceived by the users. Consequently, this paper examines the most recent research devoted on public opinion data mining. On the basis of extensive literature review, we outline the latest developments and trends in the field of public opinion data mining with special focus on sentiment analysis. Our main goal is to provide a self-contained comprehensive summary that might be used as a basis for design and development of AI systems aimed to mine the public opinion. [ABSTRACT FROM AUTHOR]

Depression is the most prevailing mental health problem nowadays and researchers expect it to deepen especially in most developed countries due to constantly aging population. This development would pose extreme financial burden on public finance. The SHARE project provides a huge variety of data sets including a mental health module used to form the 12-item EURO-D scale, which serves as a common metrics for depressions. However, in wave 7 this module includes respondents from only 12 countries. Consequently, this paper aims to construct data-driven indices reflecting the level of depression in Europe. We employ factor analysis and as a result, we derive two latent variables. High positive values of the first one indicate lack of depression, while high positive values of the second variable might be indicative for presence of serious depressive symptoms. We then inspect cross-national differences in terms of these two latent variables. Our major findings suggest that geographic location tends to have some effect on the presence of depressive symptoms among elderly people in Europe. [ABSTRACT FROM AUTHOR]

Aging population across developed countries is a problem that no nation has faced before. It puts severe financial burden on every public budget. Non-financial burdens are also common. Depression turns out to be the most common mental disease. Studying the features, that could boost the exhibition of depressive symptoms, and those, that could protect individuals from them, would have significant effect on the precision of public health policies. That is the aim of the paper. We base our research on wave 7 data of the Survey of Health, Aging, and Retirement in Europe that for the first time summarizes information about 27 European countries and Israel. As a preliminary step in our analysis, we construct two latent variables used to quantify the level of depression. On the basis of extensive literature review, we identify a long list of potential features and apply lasso regression so as to select those of them that contribute most in explaining the variance of each of the latent variables. Delivered results indicate that among the most common features, leading to poor mental health, are working conditions, self-perceived health and the presence of functional limitations. On the other hand, educational attendance, lack of chronic diseases and sport activities are among the factors that protect from exhibiting depression symptoms. [ABSTRACT FROM AUTHOR]

Many recent papers conclude on short-term profitable patterns based on trading strategies relying on the belief that best (worst) performing securities over the past short-term period tend to continue to perform well (poorly) over the subsequent period of up to 12 months. Along with exploiting past price patterns, ever since the early 1980s the researchers' attention is drawn on the usefulness of company financial statements in predicting stock market returns. We argue on the necessity to develop a framework that would enable investors to support their everyday decisions by taking into account information retrieved from financial statements in an automated manner. We propose an ML-based AI system that inputs historic price information as well as information on a number of balance sheet items, profit and loss items, and cash flow items for the company of interest and predicts the probability of its stock price going down in the next period. The prediction relies on a dynamic automated selection of significant accounting and historic features. Tracking the dynamics in the structure of significant features could shed additional light on company specific issues that might be taken into account by investors when making their decisions. [ABSTRACT FROM AUTHOR]

In a technology-rich business environment, accounting educators are challenged to develop the required skill set in accounting graduates and to prepare them for a successful career in a data-driven environment. As the deployment of AI and ML advances may substantially improve almost every area of accounting, future professionals have to be well equipped with the needed knowledge and competences to benefit from their capabilities. Universities are searching for innovative approaches and good practices for implementing AI and ML techniques into accounting curriculum. This paper introduces an AI-based case study for accounting students, supported by clear learning objectives, methods and detailed implementation guidance. By playing their assigned role as company management's consultants under the case hypothetical scenario, students have to complete several individual and group tasks that demonstrate how the quality of investment decisions might be improved by enhancing the usefulness of companies' financial statements. Students are also encouraged to develop their analytical thinking and interpretation skills as well as their communication and deliberation proficiencies complementing the targeted competences in AI and ML. [ABSTRACT FROM AUTHOR]

Momentum trading strategies assume that past winners, i.e. the best performing equity securities over the past up to 12 months tend to continue to perform well over the subsequent short-term period. This paper examines the momentum effect for the New York Stock Exchange over a period of more than 20 years. Although this issue is extensively studied, the volatility structure of momentum profits is still unexplored. Therefore, our main goal is to shed light on the hidden temporal patterns in the raw profits generated by winners trading strategies. For this purpose, we use continuous wavelet analysis so as to decompose the observed volatility of raw profit series over a range of investment horizons. We discover clearly pronounced patterns in times of turmoil that could not be explained by application of time domain filtering. Furthermore, we identify which investment horizons impact most heavily the dynamics in momentum profits during the Dot-com bubble and during the Great Recession. We consider this breakdown as particularly useful when studying changes in crowd investors’ behavior and beliefs. [ABSTRACT FROM AUTHOR]

Momentum effect has been widely studied for almost three decades by both behavioral and conventional scientists. It indicates the presence of return predictability on international investment markets across different asset classes, which contradicts to the Efficient Market Hypothesis. Stocks, that have performed best (worst) over the previous short-run, tend to continue to earn high (poor) return over the subsequent up to twelve months. This paper examines the profitability of winners-only momentum trading strategies on the New York Stock Exchange (NYSE) and on the Shanghai Stock Exchange (SHSE) over the period between 1-July-1997 and 31-December-2018. We estimate the performance of 64 momentum trading strategies. The article analyses the relationship between the documented momentum effect on the two capital markets and the cultural characteristics for the corresponding societies on the basis of the 6-Dimensions Culture Model by Hofstede. Outlined are implications for the relation between the rising of momentum effect and cultural specifics. Our empirical results show, that for the overall analyzed sample period there is return predictability, earning about 1% per week for both stock exchanges with some dominancy for the NYSE. [ABSTRACT FROM AUTHOR]