CARMA 2024 - 6th International Conference on Advanced Research Methods and Analytics

Research methods in economics and social sciences are evolving with the increasing availability of Internet and Big Data sources of information. As these sources, methods, and applications become more interdisciplinary, the 6th International Conference on Advanced Research Methods and Analytics (CARMA) is a forum for researchers and practitioners to exchange ideas and advances on how emerging research methods and sources are applied to different fields of social sciences as well as to discuss current and future challenges.

URI permanente para esta colecciónhttps://riunet.upv.es/handle/10251/207246

Examinar

Mostrando 1 - 20 de 46

Access status: Abierto ,
Challenges in Upholding Human Autonomy through the Right to be Forgotten
(Editorial Universitat Politècnica de València, 2024-07-16) Zarrin, Sadaf; Unceta Mendieta, Irene
[EN] The paper examines the difficulties and challenges in implementing the right to be forgotten, highlighting the importance of this right for individual autonomy and privacy. It explores the main obstacles to upholding this right from legal, ethical, practical, and technical viewpoints, providing a summary of the existing problems and making recommendations for potential solutions. To improve the applicability of human rights in the digital world, the research emphasizes the significance of public awareness, international collaboration, and improvements in machine unlearning solutions. In order to assist the effective application of the right to be forgotten, the paper ends with suggestions for future research. These ideas seek to achieve a balance between autonomy, the need for privacy, and the rapid development of technology in digital spaces
Access status: Abierto ,
Eliciting and Retrieving the Feedback-Loop. Exploring Elicitation Interview Techniques for Detecting Algorithmic Feedback on Social Media and Cultural Consumption
(Editorial Universitat Politècnica de València, 2024-07-16) Punziano, Gabriella; Gandini, Alessandro; Caliandro, Alessandro; Airoldi, Massimo; Padricelli, Giuseppe; Acampa, Suania; Trezza, Domenico; Crescentini, Noemi; Rama, Ilir; European Commission; Ministero dell'Università e della Ricerca
[EN] This article introduces elicitative interviewing techniques in the context of algorithmic feedback detection on social media about cultural consumption. This article presents elicitation interviewing methods to identify algorithmic feedback concerning cultural consumption on social media. The initial section will clarify the notion of influence in algorithm-driven consumption decisions on these platforms. The second part will underscore the necessity for finely nuanced qualitative methodologies to dissect the conceptual facets essential for analysis within such contexts of influence and dynamics. The main interviewing techniques for finalizing data collection with this intent will then be reviewed. The third part will present an example of a survey instrument that uses the elicitation component to achieve the essence of the feedback-loop between algorithms and cultural consumption choices that underlie the PRIN ALGOFEED survey. Finally, this detection phase's placement within the project and its role as an enhancer of the preceding collection and analysis stages will be elucidated, emphasizing the benefits of this decision and the potential pitfalls that necessitate proper attention and scrutiny.
Access status: Abierto ,
Towards Intangible Value Quantification: Scope, Limits & Shortages of Artificial Intelligence applications
(Editorial Universitat Politècnica de València, 2024-07-16) Domínguez Gil, Salvador; San José Cabrero, Andrea; Sánchez Gea, Antonio; Miguel-Sin, Pilar; Ramírez Pacheco, Gema
[EN] The application of Artificial Intelligence (AI) in the realm of economic markets, particularly in the business and real estate sectors, has witnessed substantial growth. However, its effectiveness is curtailed by several limitations, especially in the context of the rising valuation of intangible assets. The intangible nature of assets such as brand value, environmental impact or social impact, among others, presents a challenge for AI, which relies on quantifiable data for analysis and decision-making. The intrinsic volatility and uncertainty of markets, heightened by the intangible asset valuation, further complicate the AI's predictive accuracy and adaptability throughout the time.AI models, primarily dependent on historical data, struggle to accurately forecast market movements influenced by intangible factors, which are often subjective and dynamically changing. This limitation is particularly pronounced in the real estate promotion sector, where the perceived value of properties can be significantly affected by intangible elements like location prestige or architectural uniqueness. Additionally, the ethical implications of AI deployment, such as data privacy concerns and potential biases in algorithmic decision-making, pose further constraints on its application in these sectors. While AI offers transformative potential for economic markets, its current limitations in handling the valuation of intangibles, market volatility, and ethical considerations necessitate a cautious and complementary approach to its integration into business and real estate promotion strategies, specially in the concern of life-cycle approaches.
Access status: Abierto ,
Google trends forecasting of youth employment
(Editorial Universitat Politècnica de València, 2024-07-16) Bruijn, Nathan; Wijnhoven, Fons; Effing, Robin
[EN] The forecasting field has been using the surge in big data and advanced computational capabilities. This article discusses the methodological issues of Google Trends (GT) data reliability and forecasting validity for youth unemployment forecasts. We demonstrate the problems with static GT forecasting procedures and show a 44% increase in forecasting accuracy by applying time-varying model respecification forecasting.
Access status: Abierto ,
Unveiling New Insights From Textual Unstructured Big Data in Politics Through Deep Learning
(Editorial Universitat Politècnica de València, 2024-07-16) Caliskan, Ufuk; Pappagallo, Angela; Ortame, Francesco; Bruno, Mauro; Pugliese, Francesco
[EN] Over the past decade, social media platforms have undergone significant and rapid expansion. One of the key challenges has been effectively analysing the vast amount of unstructured user-generated data they produce. This research delves into the analysis of Italian Twitter data through the application of advanced deep learning models across three primary objectives: text classification, sentiment analysis, and hate analysis. Five cutting-edge models are evaluated, each utilizing distinct word embeddings.Furthermore, this study investigates the effects of processing emojis and emoticons in Italian tweets on sentiment and hate analysis. We compare model performances and suggest optimized approaches for each task. Finally, we apply these methodologies to real-world Twitter data and present our findings through multiple graphs and statistical analyses. This study demonstrates the possibility of extracting new insights and novel information from unstructured textual Big Data in Politics.
Access status: Abierto ,
Nowcasting food insecurity interest Google Trends data
(Editorial Universitat Politècnica de València, 2024-07-16) Caravaggio, Nicola; Carneiro, Bia; Resce, Giuliano
[EN] This research explores the potential of Google Trends (GT) data as a tool for generating a daily index of food insecurity at the national level, focusing on regions monitored by the Famine Early Warning Systems Network (FEWS NET) and the Global Fragility Act (GFA). Drawing inspiration from previous studies on GT's predictive capabilities, the authors employ Natural Language Processing (NLP) to analyse food security reporting from FEWS NET documents. We identify key predictors of food insecurity using a LASSO regression approach and construct a daily economic sentiment index (DESI) for each country. Unlike traditional methods, the study considers multiple languages and weights search terms based on LASSO coefficients. The resulting Synthetic Search Interest (SSI) index for food insecurity demonstrates a statistically significant correlation with FAO's share of the population in severe food insecurity, affirming GT's potential as a monitoring tool. The research contributes a novel methodology and insights into leveraging real-time data for early warnings in food security.
Access status: Abierto ,
Work Realities and Behavioral Risk Factors in Italy
(Editorial Universitat Politècnica de València, 2024-07-16) Andreella, Angela; Campostrini, Stefano
[EN] The connection between health, work environment, and job characteristics is a relevant issue in public health. However, it is often underexplored due to a lack of reliable data. To address this gap, we have delved into the subject using data from an NCDs-risk factor surveillance system (PASSI). We have examined information collected from respondents regarding their occupations relating to risk factors and health status. The proposed analysis employs text mining and cluster approach for categorical variables to identify sub-populations characterized by different socio-economic situations, risk factors, and job types. Although further analyses are needed to explore the potential of this approach better, initial results are promising. They highlight the practical implications of our findings for public health policies. For example, we found that occupations related to the building industry (for males) and healthcare professions (for females) appear to be associated with higher behavioral risk factors, which could inform targeted interventions.
Access status: Abierto ,
Electoral abstention and information sources among undergraduate university students
(Editorial Universitat Politècnica de València, 2024-07-16) Mora Rojo, Jorge; Tomás, José Manuel; Yeste, Víctor; Cebrián, Eduardo
[EN] A quantitative research study was conducted to examine electoral abstention among undergraduate students in Valencia during Spain’s General Elections held in July 2023. Data were collected through a survey based on a questionnaire designed specifically for this purpose, focusing on the electoral behavior, socioeconomic and demographic profiles, and information sources. A multivariate statistic model was used to explain the likelihood of students non-voting. Key findings indicate that abstention it is not related to a demographic profile. However, a negative correlation with abstention was found among information sources such as party websites/social media, YouTube, printed press and TV, whereas a positive relationship was found with the use of blogs and forums.
Access status: Abierto ,
From Crisis to Opportunity: A Google Trends Analysis of Global Interest in Distance Education Tools During and Post the COVID- 19 Pandemic
(Editorial Universitat Politècnica de València, 2024-07-16) Talagala, Priyanga; Talagala, Thiyanga
[EN] This study investigated the impact of COVID-19 on global attention towards different distance education tools. We used Google Trend search queries as a proxy to quantify the popularity and public interest in different distance education solutions under 11 sub-segments, which include collaboration platforms, online proctoring, and resources for psychosocial support. The study employs both visual and analytical approaches to analyse global web search queries during and post the COVID-19 pandemic. Through cross-correlation analysis and dynamic time-warping analysis, the study confirms the contemporaneous and lead-lag relationships between COVID-19 and distance education-related search terms. Furthermore, the study highlights the critical role of psychosocial support in promoting the well-being of students and teachers during a pandemic. The study emphasizes the importance of Google footprint analysis in determining the most popular online education resources designed for different educational goals. This feature allows educators to gain insight into prominent distant education options, boosting their online teaching.
Access status: Abierto ,
The use of non-official data source for the analysis of public events: evidences from the Eurovision Song Contest 2022
(Editorial Universitat Politècnica de València, 2024-07-16) Forciniti, Alessia; Marletta, Andrea; Moretti, Magda
[EN] The use of non-official data sources as Twitter has been implemented for the monitoring of social and public events in many different fields during last years. Following this issue, this work proposes to analyse a very well-known musical event, the Eurovision Song Contest (ESC) 2022 using tweets pooled by the official hashtag of the competition. From a methodological point of view, text mining techniques have been applied to detect the most influencing terms and topics tweeted by users during the show and to compare the official results of the contest with a ranking only based on the appreciation of the Twitter users on posts relative to the participant countries.
Access status: Abierto ,
A Bibliometric Study of Stakeholder Opinion Mining and Sentiment Analysis in Crisis Communication
(Editorial Universitat Politècnica de València, 2024-07-16) Molavi, Homa; Zhang, Lihong
[EN] In the contemporary landscape, the ability to effectively manage crises and communicate with stakeholders is paramount for organizations. As the frequency and complexity of crises continue to escalate, understanding stakeholder opinions and sentiments becomes increasingly crucial for crafting timely and appropriate responses. This bibliometric study delves into the landscape of stakeholder opinion mining and sentiment analysis within crisis communication, aiming to discern trends, identify key contributors, and uncover potential gaps in the existing literature. Leveraging data from the Scopus database from 2004 to 2024, the analysis reveals a notable increase in publications over time, particularly since 2019, highlighting the growing interest in this field. The United States, the United Kingdom, and Germany emerge as leading contributors, while institutions such as The University of Texas at Austin and Universiteit van Amsterdam demonstrate significant productivity. However, limited collaboration between top institutions and authors suggests opportunities for enhanced knowledge exchange and interdisciplinary collaboration.
Access status: Abierto ,
Potential of ChatGPT in predicting stock market trends based on Twitter Sentiment Analysis
(Editorial Universitat Politècnica de València, 2024-07-16) Mumtaz, Ummara; Mumtaz, Summaya
[EN] The rise of ChatGPT has brought a notable shift to the AI sector, with its exceptional conversational skills and deep grasp of language. Recognizing its value across different areas, our study investigates ChatGPT's capacity to predict stock market movements using only social media tweets and sentiment analysis. We aim to see if ChatGPT can tap into the vast sentiment data on platforms like Twitter to offer insightful predictions about stock trends. We focus on determining if a tweet has a positive, negative, or neutral effect on two big tech giants Microsoft and Google’s stock value. Our findings highlight a positive link between ChatGPT's evaluations and the following day's stock results for both tech companies. This research enriches our view on ChatGPT's adaptability and emphasizes the growing importance of AI in shaping financial market forecasts.
Access status: Abierto ,
Bibliometrics and Scientometrics of the Business Agility
(Editorial Universitat Politècnica de València, 2024-07-16) Lesníková, Petra; Janakova Sujova, Andrea; Vedecká Grantová Agentúra MSVVaS SR a SAV
[EN] Bibliometric analysis is an important tool in scientific research designed to explore and analyse a range of scientific data. The purpose of this paper is to highlight the growing importance and relevance of business agility issues in the scientific community. The aim is to provide a brief insight into business agility through bibliometric analysis of articles included in the WOS and Scopus databases. As a result, a comparison of these databases is presented along with a description of the resulting clusters using the software tools VOSviewer and SciMAT. The area of interest in the databases is Business, Economics, Management and Finance in the publication years 1994-2023. The results show that although the databases overlap to some extent, there are some slight differences in terms of bibliometrics or scientometrics. Although the Scopus database had a higher number of publications, the number of keyword occurrences is higher in the database WOS. There are also slight differences in the most numerous keywords. In terms of clusters, the number is the same, but slight differences are also observed. Based on the analysis of the occurrence of keywords, it is possible to note an increased interest in the issue of agility, which is linked to a number of other areas of management. The Scopus database is recommended to study business agility.
Access status: Abierto ,
Data-Driven Strategies for Early Detection of Corporates’ Financial Distress
(Editorial Universitat Politècnica de València, 2024-07-16) Riccio, Donato; Bifulco, Giuseppe; Francesco, Paolone; Mazzitelli, Andrea; Maturo, Fabrizio
[EN] Scholars have taken a keen interest in predicting corporate crises in the past decades. However, most studies focused on classical parametric models that, by their nature, can consider few predictors and interactions and must respect numerous assumptions. Over the past few years, the economy has faced a severe structural crisis that has resulted in significantly lower income, cash, and capital levels than in the past. This crisis has led to insolvency and bankruptcy in many cases. Hence, there is a renewed interest in research for new models for forecasting business crises using novel advanced statistical learning techniques. The study shows that using tree-based methods and hyper-parameters optimization leads to excellent results in terms of accuracy. Moreover, this approach allows us to automatically consider all possible interactions and discover relevant aspects never considered in past studies. This line of research provides fascinating results that can bring new knowledge into the reference literature.
Access status: Abierto ,
Digitalisation - the Basis for Building an Agile Enterprise
(Editorial Universitat Politècnica de València, 2024-07-16) Janáková Sujová, Andrea; Lesníková, Petra; Vedecká Grantová Agentúra MSVVaS SR a SAV
[EN] Enterprise digitalisation and agility are two key concepts that can work together to contribute to an organisation's competitiveness. Enterprise digitalization as a process of transforming traditional business models and processes using digital technologies enables the implementation of agile principles. Enterprise agility as the ability to adapt quickly and efficiently to unpredictable changes in the environment is becoming an important competitive factor. The aim of the article is to reveal the interactions between digitalization and enterprise agility and to present the results of primary research in industrial enterprises of the Slovak Republic focused on the perception of the importance of digitalization in the context of agility. The results showed that digitalization is an important element of agility and an essential starting point in building an agile enterprise. Slovak industrial enterprises consider digitisation as an important help in coping with unexpected changes such as the coronacrisis, as a result of which digitisation has accelerated. However, the current adverse global circumstances mean that one third of enterprises have reduced or stopped digitisation altogether and the number of digitising enterprises has declined over three years. Digitisation can transform business models and create agile operating models.
Access status: Abierto ,
Digital Transformation in Supply Chain Management: A Bibliometric Analysis
(Editorial Universitat Politècnica de València, 2024-07-16) Zhang, Lihong; Banihashemi, Saeed; Rui, Aiwen; Chen, Song
[EN] In an era dominated by digital advancements, Supply Chain Management (SCM) is undergoing significant transformations. This study aims to outline a digitally-enabled SCM framework by examining prevalent research themes, methodologies, collaboration models, and groundbreaking contributions. Leveraging bibliometric analysis techniques, 600 articles from 2002 to 2023, sourced from 3 top databases, was conducted using CiteSpace for keyword analysis, co-citation analysis, and emerging term evaluations. The research identifies three distinct phases in this field: initial, incremental, and accelerated, with a notable surge in research activity post-2017, particularly after 2019. China, the US, and the UK are major contributors. Dominant topics include technology integration, efficiency, globalization challenges, strategic management, and sustainability. The study reveals limited collaboration among authors but highlights influential scholars. Given the centrality of DT, it emphasizes the need for interdisciplinary exploration and consideration of rapidly evolving digital paradigms in future research endeavors.
Access status: Abierto ,
Improving Accuracy in Geospatial Information Transfer: A Population Density-Based Approach
(Editorial Universitat Politècnica de València, 2024-07-16) Pérez, Virgilio; Pavía, Jose
[EN] The R package sc2sc offers fundamental tools for transferring information between census sections and postal codes in Spain, based on the cartography of these geographic segmentations. However, certain aspects for improvement have been identified. This document presents a substantial improvement to the package, optimizing the cp2sc function, which facilitates the transfer of information from postal codes to census sections. The introduced improvement considers population density as a corrective factor in the process, resulting in a more accurate and relevant data allocation. Various use cases highlight the improvement of the new methodology, though they also underline the need to work with updated and precise cartography, opening new lines of research and future work.
Access status: Abierto ,
Economic forecasting with non-specific Google Trends sentiments: Insights from US Data
(Editorial Universitat Politècnica de València, 2024-07-16) Diaf, Sami; Schütze, Florian
[EN] The influence of specific Google Trends search queries measuring various sentiments on economic performance and stock markets has been extensively documented and used for many purposes. This paper examines the predictive power of queries measuring non-specific sentiment on key macroeconomic variables when linked to a comprehensive sentiment dictionary. The analysis shows that non-specific sentiments do not improve the forecasting quality of the US economy as a whole, except for unemployment, which was found to be predictable for all sentiments. Consequently, the authors suggest that economic-related sentiments with carefully selected words should be used in Google Trends search queries to improve predictive performance. However, if a socio-cultural analysis is to be performed, non-specific sentiments would be suggested, as they can be predicted by the real economic time series of unemployment.
Access status: Abierto ,
The potential of Google Trend in estimating the absorption rate of European structural funds
(Editorial Universitat Politècnica de València, 2024-07-16) Caravaggio, Nicola; Pierucci, Eleonora; Resce, Giuliano
[EN] This study investigates the relationship between Google Trends (GT) interest in European Structural and Investment Funds (ESIF) and the absorption rate across 27 European Union countries. Utilizing a two-way fixed effect methodology, we analyse annual GT data from 2007 to 2016. Results reveal a consistently positive and statistically significant explicative power of lagged values of GT interest on absorption rates. The findings suggest that the online search behaviour regarding ESIF correlates with fund absorption, revealing the potential predictive value of GT data in the context of regional cohesion policies. This study contributes to the literature on the practical applications of GT across diverse domains and underscores its relevance in predicting the implementation of EU cohesion policies.
Access status: Abierto ,
Prediction of SMEs Bankruptcy at the Industry Level with Balance Sheets and Website Indicators
(Editorial Universitat Politècnica de València, 2024-07-16) Bottai, Carlo; Crosato, Lisa; Liberati, Caterina
[EN] This paper addresses the importance of industry-specific models for SMEs bankruptcy prediction, building on earlier research finding larger predictive accuracy and enhanced temporal stability. Using Italian data, we propose separate bankruptcy prediction models for a few industries based on balance sheet data and explore the predictive power of SMEs' website html code structure. Our findings suggest that website data can serve as a valid complementary source for bankruptcy prediction, with different performances across sectors. We observe a certain degree of sectoral heterogeneity in the importance of financial ratios, firm-specific characteristics, and website structure, calling for an industry-tailored approach in bankruptcy prediction models.

Examinar

Envíos recientes