Explainable artificial intelligence and agile decision-making in supply chain cyber resilience

Decision Support Systems, Journal Year: 2024, Volume and Issue: 180, P. 114194 - 114194

Published: Feb. 17, 2024

Although artificial intelligence can contribute to decision-making processes, many industry players lag behind pioneering companies in utilizing intelligence-driven technologies, which is a significant problem. Explainable be viable solution mitigate this This paper proposes research model address how explainable impact processes. Using an experimental design, empirical data collected test the model. one of pioneer papers providing evidence about on supply chain We propose serial mediation path, includes transparency and agile decision-making. Findings reveal that enhances transparency, thereby significantly contributing for improving cyber resilience during cyberattacks. Moreover, we conduct post hoc analysis using text explore themes present tweets discussing decision support systems. The results indicate predominantly positive attitude towards within these Furthermore, reveals two main emphasize importance explainability, interpretability intelligence.

Language: Английский

---

An emoji feature-incorporated multi-view deep learning for explainable sentiment classification of social media reviews

Technological Forecasting and Social Change, Journal Year: 2024, Volume and Issue: 202, P. 123326 - 123326

Published: March 16, 2024

Sentiment analysis has demonstrated its value in a range of high-stakes domains. From financial markets to supply chain management, logistics, and technology legitimacy assessment, sentiment offers insights into public sentiment, actionable data, improved decision forecasting. This study contributes this growing body research by offering novel multi-view deep learning approach that incorporates non-textual features like emojis. The proposed considers both textual emoji views as distinct emotional information for the classification model, results acknowledge their individual combined contributions analysis. Comparative with baseline classifiers reveals incorporating significantly enriches analysis, enhancing accuracy, F1-score, execution time model. Additionally, employs LIME explainable provide model's decision-making process, enabling businesses understand factors driving customer sentiment. present literature on text context social media provides an innovative analytics method extract valuable from electronic word mouth (eWOM), which can help them stay ahead competition rapidly evolving digital landscape. In addition, findings paper have important implications policy development communication monitoring. Recognizing importance emojis expression inform policies helping better tailor solutions address concerns public.

Language: Английский

---

Extreme gradient boosting trees with efficient Bayesian optimization for profit-driven customer churn prediction

Technological Forecasting and Social Change, Journal Year: 2023, Volume and Issue: 198, P. 122945 - 122945

Published: Nov. 6, 2023

Language: Английский

---

The role of feature importance in predicting corporate financial distress in pre and post COVID periods: Evidence from China

International Review of Financial Analysis, Journal Year: 2023, Volume and Issue: 90, P. 102851 - 102851

Published: Aug. 6, 2023

Language: Английский

---

Volatility forecasting of crude oil futures based on Bi-LSTM-Attention model: The dynamic role of the COVID-19 pandemic and the Russian-Ukrainian conflict

Resources Policy, Journal Year: 2023, Volume and Issue: 88, P. 104319 - 104319

Published: Nov. 15, 2023

Language: Английский

---

On the prediction of systemic risk tolerance of cryptocurrencies

Technological Forecasting and Social Change, Journal Year: 2023, Volume and Issue: 198, P. 122963 - 122963

Published: Nov. 10, 2023

Language: Английский

---

Predicting M&A targets using news sentiment and topic detection

Technological Forecasting and Social Change, Journal Year: 2024, Volume and Issue: 201, P. 123270 - 123270

Published: Feb. 13, 2024

This paper uses news sentiment and topics to discuss the challenges opportunities of predicting mergers acquisition (M&A) targets. We explore effect investor on identifying M&As targets how company-specific articles can be used as a source obtain richer information various corporate events. propose framework incorporating into M&A target prediction model, utilising state-of-the-art transformer-based analysis topic modelling approaches. evaluate textual features' predictive power using real-world dataset US UK non-target companies from 2020 2021, with several experiments conducted reveal contribution thematic focus prediction. A profit-based objective function is proposed overcome inherent class imbalance problem in dataset. Our findings suggest that news-based models outperform traditional statistical single machine learning methods, indicating need for more robust less prone overfitting ensemble methods. Additionally, our study provides evidence positive negative likelihood M&A. research has important implications investors analysts who seek identify investment opportunities.

Language: Английский

---

Driving forces of digital transformation in chinese enterprises based on machine learning

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: March 14, 2024

Abstract With advanced science and digital technology, transformation has become an important way to promote the sustainable development of enterprises. However, existing research only focuses on linear relationship between a single characteristic transformation. In this study, we select data Chinese A-share listed companies from 2010 2020, innovatively use machine learning method explore differences in predictive effects multi-dimensional features enterprises based Technology-Organization-Environment (TOE) theory, thus identifying main drivers affecting fitting models with stronger effect. The study found that: first, by comparing traditional regression models, it is that prediction ability ensemble earning generally higher than tradition measurement method. For sample selected research, XGBoost LightGBM have strong explanatory high accuracy. Second, compared technical driving force environmental force, organizational greater impact. Third, among these characteristics, equity concentration executives’ knowledge level dimension greatest impact Therefore, enterprise managers should always pay attention decision-making role level. This further enriches literature enterprises, expands application economics, provides theoretical basis for enhance

Language: Английский

---

Credit risk prediction based on an interpretable three-way decision method: Evidence from Chinese SMEs

Applied Soft Computing, Journal Year: 2024, Volume and Issue: 157, P. 111538 - 111538

Published: March 26, 2024

Language: Английский

---

Practical forecasting of risk boundaries for industrial metals and critical minerals via statistical machine learning techniques

International Review of Financial Analysis, Journal Year: 2024, Volume and Issue: 94, P. 103252 - 103252

Published: March 28, 2024

Language: Английский

---