Artificial Intelligence Applications in Financial Technology

Journal Name	Impact Factor	CiteScore	Launched Year	First Decision (median)	APC
AI ai	3.1	7.2	2020	17.6 Days	CHF 1600
Big Data and Cognitive Computing BDCC	3.7	7.1	2017	18 Days	CHF 1800
Economies economies	2.1	4.0	2013	21.7 Days	CHF 1800
International Journal of Financial Studies ijfs	2.1	3.7	2013	29.4 Days	CHF 1800
Journal of Theoretical and Applied Electronic Commerce Research jtaer	5.1	9.5	2006	32 Days	CHF 1000
Sustainability sustainability	3.3	6.8	2009	20 Days	CHF 2400

20 pages, 11158 KiB

Open AccessArticle

Quantitative Stock Selection Model Using Graph Learning and a Spatial–Temporal Encoder

by Tianyi Cao, Xinrui Wan, Huanhuan Wang, Xin Yu and Libo Xu

J. Theor. Appl. Electron. Commer. Res. 2024, 19(3), 1756-1775; https://doi.org/10.3390/jtaer19030086 - 15 Jul 2024

Viewed by 1047

Abstract

In the rapidly evolving domain of finance, quantitative stock selection strategies have gained prominence, driven by the pursuit of maximizing returns while mitigating risks through sophisticated data analysis and algorithmic models. Yet, prevailing models frequently neglect the fluid dynamics of asset relationships and [...] Read more.

In the rapidly evolving domain of finance, quantitative stock selection strategies have gained prominence, driven by the pursuit of maximizing returns while mitigating risks through sophisticated data analysis and algorithmic models. Yet, prevailing models frequently neglect the fluid dynamics of asset relationships and market shifts, a gap that undermines their predictive and risk management efficacy. This oversight renders them vulnerable to market volatility, adversely affecting investment decision quality and return consistency. Addressing this critical gap, our study proposes the Graph Learning Spatial–Temporal Encoder Network (GL-STN), a pioneering model that seamlessly integrates graph theory and spatial–temporal encoding to navigate the intricacies and variabilities of financial markets. By harnessing the inherent structural knowledge of stock markets, the GL-STN model adeptly captures the nonlinear interactions and temporal shifts among assets. Our innovative approach amalgamates graph convolutional layers, attention mechanisms, and long short-term memory (LSTM) networks, offering a comprehensive analysis of spatial–temporal data features. This integration not only deciphers complex stock market interdependencies but also accentuates crucial market insights, enabling the model to forecast market trends with heightened precision. Rigorous evaluations across diverse market boards—Main Board, SME Board, STAR Market, and ChiNext—underscore the GL-STN model’s exceptional ability to withstand market turbulence and enhance profitability, affirming its substantial utility in quantitative stock selection. Full article

(This article belongs to the Topic Artificial Intelligence Applications in Financial Technology)

► Show Figures

Figure 1

18 pages, 2844 KiB

Open AccessArticle

Risk Analysis of Bankruptcy in the U.S. Healthcare Industries Based on Financial Ratios: A Machine Learning Analysis

by Hadi Gholampoor and Majid Asadi

J. Theor. Appl. Electron. Commer. Res. 2024, 19(2), 1303-1320; https://doi.org/10.3390/jtaer19020066 - 30 May 2024

Cited by 2 | Viewed by 1312

Abstract

The prediction of bankruptcy risk poses a formidable challenge in the fields of economics and finance, particularly within the healthcare industry, where it carries significant economic implications. The burgeoning field of healthcare electronic commerce, continuously evolving through technological advancements and changing regulations, introduces [...] Read more.

The prediction of bankruptcy risk poses a formidable challenge in the fields of economics and finance, particularly within the healthcare industry, where it carries significant economic implications. The burgeoning field of healthcare electronic commerce, continuously evolving through technological advancements and changing regulations, introduces additional layers of complexity. We collected financial data from 1265 U.S. healthcare industries to predict bankruptcy based on 40 financial ratios using multi-class classification machine learning models across various industry subsectors and market capitalizations. The exceptionally high post-tuning accuracy rates, exceeding 90%, along with high-performance metrics solidified the robustness and exceptional predictive capability of the gradient boosting model in bankruptcy prediction. The results also demonstrate the power and sensitivity of financial ratios in predicting bankruptcy based on financial ratios. The Altman models highlight the return on investment (ROI) as the most important parameter for predicting bankruptcy risk in healthcare industries. The Ohlson model identifies return on assets (ROA) as an important ratio specifically for predicting bankruptcy risk within industry subsectors. Furthermore, it underscores the significance of both ROA and the enterprise value to earnings before interest and taxes (EV/EBIT) ratios as important parameters for predicting bankruptcy based on market capitalization. Recognizing these ratios enables proactive decision making that enhances resilience. Our findings contribute to informed risk management strategies, allowing for better management of healthcare industries in crises like those experienced in 2022 and even on a global scale. Full article

(This article belongs to the Topic Artificial Intelligence Applications in Financial Technology)

► Show Figures

Figure 1

23 pages, 2277 KiB

Open AccessArticle

The Impact of Academic Publications over the Last Decade on Historical Bitcoin Prices Using Generative Models

by Adela Bâra and Simona-Vasilica Oprea

J. Theor. Appl. Electron. Commer. Res. 2024, 19(1), 538-560; https://doi.org/10.3390/jtaer19010029 - 6 Mar 2024

Cited by 2 | Viewed by 1905

Abstract

Since 2012, researchers have explored various factors influencing Bitcoin prices. Up until the end of July 2023, more than 9100 research papers on cryptocurrencies were published and indexed in the Web of Science Clarivate platform. The objective of this paper is to analyze [...] Read more.

Since 2012, researchers have explored various factors influencing Bitcoin prices. Up until the end of July 2023, more than 9100 research papers on cryptocurrencies were published and indexed in the Web of Science Clarivate platform. The objective of this paper is to analyze the impact of publications on Bitcoin prices. This study aims to uncover significant themes within these research articles, focusing on cryptocurrencies in general and Bitcoin specifically. The research employs latent Dirichlet allocation to identify key topics from the unstructured abstracts. To determine the optimal number of topics, perplexity and topic coherence metrics are calculated. Additionally, the abstracts are processed using BERT-transformers and Word2Vec and their potential to predict Bitcoin prices is assessed. Based on the results, while the research helps in understanding cryptocurrencies, the potential of academic publications to influence Bitcoin prices is not significant, demonstrating a weak connection. In other words, the movements of Bitcoin prices are not influenced by the scientific writing in this specific field. The primary topics emerging from the analysis are the blockchain, market dynamics, transactions, pricing trends, network security, and the mining process. These findings suggest that future research should pay closer attention to issues like the energy demands and environmental impacts of mining, anti-money laundering measures, and behavioral aspects related to cryptocurrencies. Full article

(This article belongs to the Topic Artificial Intelligence Applications in Financial Technology)

► Show Figures

Figure 1

21 pages, 2683 KiB

Open AccessFeature PaperArticle

Predicting Healthcare Mutual Fund Performance Using Deep Learning and Linear Regression

by Anuwat Boonprasope and Korrakot Yaibuathet Tippayawong

Int. J. Financial Stud. 2024, 12(1), 23; https://doi.org/10.3390/ijfs12010023 - 29 Feb 2024

Cited by 1 | Viewed by 2215

Abstract

Following the COVID-19 pandemic, the healthcare sector has emerged as a resilient and profitable domain amidst market fluctuations. Consequently, investing in healthcare securities, particularly through mutual funds, has gained traction. Existing research on predicting future prices of healthcare securities has been predominantly reliant [...] Read more.

Following the COVID-19 pandemic, the healthcare sector has emerged as a resilient and profitable domain amidst market fluctuations. Consequently, investing in healthcare securities, particularly through mutual funds, has gained traction. Existing research on predicting future prices of healthcare securities has been predominantly reliant on historical trading data, limiting predictive accuracy and scope. This study aims to overcome these constraints by integrating a diverse set of twelve external factors spanning economic, industrial, and company-specific domains to enhance predictive models. Employing Long Short-Term Memory (LSTM) and Multiple Linear Regression (MLR) techniques, the study evaluates the effectiveness of this multifaceted approach. Results indicate that incorporating various influencing factors beyond historical data significantly improves price prediction accuracy. Moreover, the utilization of LSTM alongside this comprehensive dataset yields comparable predictive outcomes to those obtained solely from historical data. Thus, this study highlights the potential of leveraging diverse external factors for more robust forecasting of mutual fund prices within the healthcare sector. Full article

(This article belongs to the Topic Artificial Intelligence Applications in Financial Technology)

► Show Figures

Figure 1

15 pages, 1569 KiB

Open AccessEditor’s ChoiceArticle

Board Expertise Background and Firm Performance

by Chiou-Yann Lee, Chun-Ru Wen and Binh Thi-Thanh-Nguyen

Int. J. Financial Stud. 2024, 12(1), 17; https://doi.org/10.3390/ijfs12010017 - 14 Feb 2024

Cited by 1 | Viewed by 2802

Abstract

This study presents a novel financial performance forecasting method that combines the threshold technique with Artificial Neural Networks (ANN). It applies the threshold regression method to identify the factors within the board of directors that influence the financial performance of traditional industries in [...] Read more.

This study presents a novel financial performance forecasting method that combines the threshold technique with Artificial Neural Networks (ANN). It applies the threshold regression method to identify the factors within the board of directors that influence the financial performance of traditional industries in Taiwan. The findings indicate that the ANN method effectively predicts financial performance by using relevant board structure data. Furthermore, the empirical results suggest that boards with more members demonstrate increased profitability. Additionally, a more significant presence of board members with accounting expertise contributes to more consistent profits. In contrast, an increased presence of members with financial expertise has a more pronounced impact on profitability. Full article

(This article belongs to the Topic Artificial Intelligence Applications in Financial Technology)

► Show Figures

Figure 1

18 pages, 2764 KiB

Open AccessArticle

Financial Anti-Fraud Based on Dual-Channel Graph Attention Network

by Sizheng Wei and Suan Lee

J. Theor. Appl. Electron. Commer. Res. 2024, 19(1), 297-314; https://doi.org/10.3390/jtaer19010016 - 2 Feb 2024

Cited by 6 | Viewed by 1896

Abstract

This article addresses the pervasive issue of fraud in financial transactions by introducing the Graph Attention Network (GAN) into graph neural networks. The article integrates Node Attention Networks and Semantic Attention Networks to construct a Dual-Head Attention Network module, enabling a comprehensive analysis [...] Read more.

This article addresses the pervasive issue of fraud in financial transactions by introducing the Graph Attention Network (GAN) into graph neural networks. The article integrates Node Attention Networks and Semantic Attention Networks to construct a Dual-Head Attention Network module, enabling a comprehensive analysis of complex relationships in user transaction data. This approach adeptly handles non-linear features and intricate data interaction relationships. The article incorporates a Gradient-Boosting Decision Tree (GBDT) to enhance fraud identification to create the GBDT–Dual-channel Graph Attention Network (GBDT-DGAN). In a bid to ensure user privacy, this article introduces blockchain technology, culminating in the development of a financial anti-fraud model that fuses blockchain with the GBDT-DGAN algorithm. Experimental verification demonstrates the model’s accuracy, reaching 93.82%, a notable improvement of at least 5.76% compared to baseline algorithms such as Convolutional Neural Networks. The recall and F1 values stand at 89.5% and 81.66%, respectively. Additionally, the model exhibits superior network data transmission security, maintaining a packet loss rate below 7%. Consequently, the proposed model significantly outperforms traditional approaches in financial fraud detection accuracy and ensures excellent network data transmission security, offering an efficient and secure solution for fraud detection in the financial domain. Full article

(This article belongs to the Topic Artificial Intelligence Applications in Financial Technology)

► Show Figures

Figure 1

19 pages, 2609 KiB

Open AccessArticle

Enhancing the Prediction of Stock Market Movement Using Neutrosophic-Logic-Based Sentiment Analysis

by Bassant A. Abdelfattah, Saad M. Darwish and Saleh M. Elkaffas

J. Theor. Appl. Electron. Commer. Res. 2024, 19(1), 116-134; https://doi.org/10.3390/jtaer19010007 - 12 Jan 2024

Cited by 6 | Viewed by 2842

Abstract

Social media platforms have allowed many people to publicly express and disseminate their opinions. A topic of considerable interest among researchers is the impact of social media on predicting the stock market. Positive or negative feedback about a company or service can potentially [...] Read more.

Social media platforms have allowed many people to publicly express and disseminate their opinions. A topic of considerable interest among researchers is the impact of social media on predicting the stock market. Positive or negative feedback about a company or service can potentially impact its stock price. Nevertheless, the prediction of stock market movement using sentiment analysis (SA) encounters hurdles stemming from the imprecisions observed in SA techniques demonstrated in prior studies, which overlook the uncertainty inherent in the data and consequently directly undermine the credibility of stock market indicators. In this paper, we proposed a novel model to enhance the prediction of stock market movements using SA by improving the process of SA using neutrosophic logic (NL), which accurately classifies tweets by handling uncertain and indeterminate data. For the prediction model, we use the result of sentiment analysis and historical stock market data as input for a deep learning algorithm called long short-term memory (LSTM) to predict the stock movement after a specific number of days. The results of this study demonstrated a predictive accuracy that surpasses the accuracy rate of previous studies in predicting stock price fluctuations when using the same dataset. Full article

(This article belongs to the Topic Artificial Intelligence Applications in Financial Technology)

► Show Figures

Figure 1

24 pages, 399 KiB

Open AccessArticle

Optimized Financial Planning: Integrating Individual and Cooperative Budgeting Models with LLM Recommendations

by I. de Zarzà, J. de Curtò, Gemma Roig and Carlos T. Calafate

AI 2024, 5(1), 91-114; https://doi.org/10.3390/ai5010006 - 25 Dec 2023

Cited by 4 | Viewed by 5972

Abstract

In today’s complex economic environment, individuals and households alike grapple with the challenge of financial planning. This paper introduces novel methodologies for both individual and cooperative (household) financial budgeting. We firstly propose an optimization framework for individual budget allocation, aiming to maximize savings [...] Read more.

In today’s complex economic environment, individuals and households alike grapple with the challenge of financial planning. This paper introduces novel methodologies for both individual and cooperative (household) financial budgeting. We firstly propose an optimization framework for individual budget allocation, aiming to maximize savings by efficiently distributing monthly income among various expense categories. We then extend this model to households, wherein the complexity of handling multiple incomes and shared expenses is addressed. The cooperative model prioritizes not only maximized savings but also the preferences and needs of each member, fostering a harmonious financial environment, whether they are short-term needs or long-term aspirations. A notable innovation in our approach is the integration of recommendations from a large language model (LLM). Given its vast training data and potent inferential capabilities, the LLM provides initial feasible solutions to our optimization problems, acting as a guiding beacon for individuals and households unfamiliar with the nuances of financial planning. Our preliminary results indicate that the LLM-recommended solutions result in budget plans that are both economically sound, meaning that they are consistent with established financial management principles and promote fiscal resilience and stability, and aligned with the financial goals and preferences of the concerned parties. This integration of AI-driven recommendations with econometric models, as an instantiation of an extended coevolutionary (EC) theory, paves the way for a new era in financial planning, making it more accessible and effective for a wider audience, as we propose an example of a new theory in economics where human behavior can be greatly influenced by AI agents. Full article

(This article belongs to the Topic Artificial Intelligence Applications in Financial Technology)

► Show Figures

Figure 1

25 pages, 412 KiB

Open AccessArticle

The Impact of Artificial Intelligence Disclosure on Financial Performance

by Fadi Shehab Shiyyab, Abdallah Bader Alzoubi, Qais Mohammad Obidat and Hashem Alshurafat

Int. J. Financial Stud. 2023, 11(3), 115; https://doi.org/10.3390/ijfs11030115 - 14 Sep 2023

Cited by 6 | Viewed by 16882

Abstract

This study determines to what extent Jordanian banks refer to and use artificial intelligence (AI) technologies in their operation process and examines the impact of AI-related terms disclosure on financial performance. Content analysis is used to analyze the spread of AI and related [...] Read more.

This study determines to what extent Jordanian banks refer to and use artificial intelligence (AI) technologies in their operation process and examines the impact of AI-related terms disclosure on financial performance. Content analysis is used to analyze the spread of AI and related information in the annual report textual data. Based on content analysis and regression analysis of data from 115 annual reports for 15 Jordanian banks listed in the Amman Stock Exchange for the period 2014 to 2021, the study reveals a consistent increase in the mention of AI-related terms disclosure since 2014. However, the level of AI-related disclosure remains weak for some banks, suggesting that Jordanian banks are still in the early stages of adopting and implementing AI technologies. The results indicate that AI-related keywords disclosure has an influence on banks’ financial performance. AI has a positive effect on accounting performance in terms of ROA and ROE and a negative impact on total expenses, which supports the dominant view that AI improves revenue and reduces cost and is also consistent with past literature findings. This study contributes to the growing body of AI literature, specifically the literature on AI voluntary disclosure, in several aspects. First, it provides an objective measure of the uses of AI by formulating an AI disclosure index that captures the status of AI adoption in practice. Second, it provides insights into the relationship between AI disclosure and financial performance. Third, it supports policymakers’, international authorities’, and supervisory organizations’ efforts to address AI disclosure issues and highlights the need for disclosure guidance requirements. Finally, it provides a contribution to banking sector practitioners who are transforming their operations using AI mechanisms and supports the need for more AI disclosure and informed decision making in a manner that aligns with the objectives of financial institutions. Full article

(This article belongs to the Topic Artificial Intelligence Applications in Financial Technology)

17 pages, 674 KiB

Open AccessArticle

Enhancing Financial Fraud Detection through Addressing Class Imbalance Using Hybrid SMOTE-GAN Techniques

by Patience Chew Yee Cheah, Yue Yang and Boon Giin Lee

Int. J. Financial Stud. 2023, 11(3), 110; https://doi.org/10.3390/ijfs11030110 - 5 Sep 2023

Cited by 7 | Viewed by 3742

Abstract

The class imbalance problem in finance fraud datasets often leads to biased prediction towards the nonfraud class, resulting in poor performance in the fraud class. This study explores the effects of utilizing the Synthetic Minority Oversampling TEchnique (SMOTE), a Generative Adversarial Network (GAN), [...] Read more.

The class imbalance problem in finance fraud datasets often leads to biased prediction towards the nonfraud class, resulting in poor performance in the fraud class. This study explores the effects of utilizing the Synthetic Minority Oversampling TEchnique (SMOTE), a Generative Adversarial Network (GAN), and their combinations to address the class imbalance issue. Their effectiveness was evaluated using a Feed-forward Neural Network (FNN), Convolutional Neural Network (CNN), and their hybrid (FNN+CNN). This study found that regardless of the data generation techniques applied, the classifier’s hyperparameters can affect classification performance. The comparisons of various data generation techniques demonstrated the effectiveness of the hybrid SMOTE and GAN, including SMOTified-GAN, SMOTE+GAN, and GANified-SMOTE, compared with SMOTE and GAN. The SMOTified-GAN and the proposed GANified-SMOTE were able to perform equally well across different amounts of generated fraud samples. Full article

(This article belongs to the Topic Artificial Intelligence Applications in Financial Technology)

► Show Figures

Figure 1

20 pages, 2261 KiB

Open AccessArticle

Predicting the Price of Bitcoin Using Sentiment-Enriched Time Series Forecasting

by Markus Frohmann, Manuel Karner, Said Khudoyan, Robert Wagner and Markus Schedl

Big Data Cogn. Comput. 2023, 7(3), 137; https://doi.org/10.3390/bdcc7030137 - 31 Jul 2023

Cited by 9 | Viewed by 10315

Abstract

Recently, various methods to predict the future price of financial assets have emerged. One promising approach is to combine the historic price with sentiment scores derived via sentiment analysis techniques. In this article, we focus on predicting the future price of Bitcoin, which [...] Read more.

Recently, various methods to predict the future price of financial assets have emerged. One promising approach is to combine the historic price with sentiment scores derived via sentiment analysis techniques. In this article, we focus on predicting the future price of Bitcoin, which is currently the most popular cryptocurrency. More precisely, we propose a hybrid approach, combining time series forecasting and sentiment prediction from microblogs, to predict the intraday price of Bitcoin. Moreover, in addition to standard sentiment analysis methods, we are the first to employ a fine-tuned BERT model for this task. We also introduce a novel weighting scheme in which the weight of the sentiment of each tweet depends on the number of its creator’s followers. For evaluation, we consider periods with strongly varying ranges of Bitcoin prices. This enables us to assess the models w.r.t. robustness and generalization to varied market conditions. Our experiments demonstrate that BERT-based sentiment analysis and the proposed weighting scheme improve upon previous methods. Specifically, our hybrid models that use linear regression as the underlying forecasting algorithm perform best in terms of the mean absolute error (MAE of 2.67) and root mean squared error (RMSE of 3.28). However, more complicated models, particularly long short-term memory networks and temporal convolutional networks, tend to have generalization and overfitting issues, resulting in considerably higher MAE and RMSE scores. Full article

(This article belongs to the Topic Artificial Intelligence Applications in Financial Technology)

► Show Figures

Figure 1

21 pages, 1173 KiB

Open AccessArticle

Explaining Policyholders’ Chatbot Acceptance with an Unified Technology Acceptance and Use of Technology-Based Model

by Jorge de Andrés-Sánchez and Jaume Gené-Albesa

J. Theor. Appl. Electron. Commer. Res. 2023, 18(3), 1217-1237; https://doi.org/10.3390/jtaer18030062 - 7 Jul 2023

Cited by 10 | Viewed by 4557

Abstract

Conversational robots powered by artificial intelligence (AI) are intensively implemented in the insurance industry. This paper aims to determine the current level of acceptance among consumers regarding the use of conversational robots for interacting with insurers and seeks to identify the factors that [...] Read more.

Conversational robots powered by artificial intelligence (AI) are intensively implemented in the insurance industry. This paper aims to determine the current level of acceptance among consumers regarding the use of conversational robots for interacting with insurers and seeks to identify the factors that influence individuals’ behavioral intention to engage with chatbots. To explain behavioral intention, we tested a structural equation model based on the Unified Theory of Acceptance and Use of Technology (UTAUT) model. It was supposed that behavioral intention is influenced by performance expectancy, effort expectancy, social influence, and trust, and by the moderating effect of insurance literacy on performance expectancy and effort expectancy. The study reveals a significant overall rejection of robotic technology among respondents. The technology acceptance model tested demonstrates a strong ability to fit the data, explaining nearly 70% of the variance in behavioral intention. Social influence emerges as the most influential variable in explaining the intention to use conversational robots. Furthermore, effort expectancy and trust significantly impact behavioral intention in a positive manner. For chatbots to gain acceptance as a technology, it is crucial to enhance their usability, establish trust, and increase social acceptance among users. Full article

(This article belongs to the Topic Artificial Intelligence Applications in Financial Technology)

► Show Figures

Figure 1

28 pages, 581 KiB

Open AccessArticle

Financial Technology Development and Green Total Factor Productivity

by Wentao Hu and Xiaoxiao Li

Sustainability 2023, 15(13), 10309; https://doi.org/10.3390/su151310309 - 29 Jun 2023

Cited by 7 | Viewed by 2499

Abstract

As a new product resulting from the deep integration of the financial industry and artificial intelligence (AI) technology, financial technology (fintech) has a significant impact on the progress of green total factor productivity (GTFP). Based on city-level data from 2011 to 2021 in [...] Read more.

As a new product resulting from the deep integration of the financial industry and artificial intelligence (AI) technology, financial technology (fintech) has a significant impact on the progress of green total factor productivity (GTFP). Based on city-level data from 2011 to 2021 in China, this paper used the super-efficiency SBM model with embedded non-expected output and the GML index method to measure the GTFP levels of 283 prefecture-level and above cities and to empirically test the impact of fintech on GTFP and its underlying mechanisms. The empirical results showed that the development of fintech had significantly promoted the improvement of GTFP, and the effect was dynamically stable. Specifically, fintech had a stronger and more significant incentive effect on GTFP in its more mature stage of development. By decomposing fintech into two dimensions, it was found that the depth of fintech development had a stronger impact on GTFP with dynamic superimposed characteristics. Mechanism analysis showed that fintech development can drive the progress of GTFP by improving resource allocation efficiency, optimizing human capital, and incentivizing technological innovation channels. Moderating effect analysis revealed that financial regulation and environmental regulation have a positive moderating effect on the baseline relationship between fintech and GTFP. Further research found that the moderating effects of financial regulation and environmental regulation exhibit significant nonlinear threshold characteristics, and the driving effect of fintech on GTFP can only reach its maximum when both are within the optimal range. This study provides valuable insights for the development and optimization of fintech, the green transformation of the real economy, and high-quality development. Full article

(This article belongs to the Topic Artificial Intelligence Applications in Financial Technology)

► Show Figures

Figure 1

17 pages, 2822 KiB

Open AccessArticle

Comparing the Performance of Corporate Bankruptcy Prediction Models Based on Imbalanced Financial Data

by Seol-Hyun Noh

Sustainability 2023, 15(6), 4794; https://doi.org/10.3390/su15064794 - 8 Mar 2023

Cited by 6 | Viewed by 2490

Abstract

Forecasts of corporate defaults are used in various fields across the economy. Several recent studies attempt to forecast corporate bankruptcy using various machine learning techniques. We collected financial information on 13 variables of 1020 companies listed on the KOSPI and KOSDAQ to capture [...] Read more.

Forecasts of corporate defaults are used in various fields across the economy. Several recent studies attempt to forecast corporate bankruptcy using various machine learning techniques. We collected financial information on 13 variables of 1020 companies listed on the KOSPI and KOSDAQ to capture the possibility of corporate bankruptcy. We propose a data processing method for small-sample domestic corporate financial data. We investigate the case of random sampling of non-bankrupt companies versus sampling non-bankrupt companies based on approximate entropy and optimized threshold based on AUC to address the imbalance between the number of bankrupt companies and the number of non-bankrupt companies. We compare the performance measures of corporate bankruptcy prediction models for the small sample data structured in two ways and the full dataset. The experimental results of this study contribute to the selection of an appropriate corporate bankruptcy prediction model. Full article

(This article belongs to the Topic Artificial Intelligence Applications in Financial Technology)

► Show Figures

Figure 1

15 pages, 705 KiB

Open AccessArticle

GALSTM-FDP: A Time-Series Modeling Approach Using Hybrid GA and LSTM for Financial Distress Prediction

by Amal Al Ali, Ahmed M. Khedr, Magdi El Bannany and Sakeena Kanakkayil

Int. J. Financial Stud. 2023, 11(1), 38; https://doi.org/10.3390/ijfs11010038 - 21 Feb 2023

Cited by 9 | Viewed by 3015

Abstract

Despite the obvious benefits and growing popularity of Machine Learning (ML) technology, there are still concerns regarding its ability to provide Financial Distress Prediction (FDP). An accurate FDP model is required to avoid financial risk at the lowest possible cost. However, in the [...] Read more.

Despite the obvious benefits and growing popularity of Machine Learning (ML) technology, there are still concerns regarding its ability to provide Financial Distress Prediction (FDP). An accurate FDP model is required to avoid financial risk at the lowest possible cost. However, in the Internet era, financial data are exploding, and they are being coupled with other kinds of risk data, making an FDP model challenging to operate. As a result, researchers presented several novel FDP models based on ML and Deep Learning. Time series data is are important to reflect the multi-source and heterogeneous aspects of financial data. This paper gives insight into building a time-series model and forecasting distress far in advance of its occurrence. To build an efficient FDP model, we provide a hybrid model (GALSTM-FDP) that incorporates LSTM and GA. Unlike other previous studies, which established models that predicted distress probability only within one year, our approach predicts distress two years ahead. This research integrates GA with LSTM to find the optimum hyperparameter configuration for LSTM. Using GA, we focus on optimizing architectural aspects for modeling the optimal network based on prediction accuracy. The results showed that our algorithm outperforms other state-of-the-art methods in terms of predictive accuracy. Full article

(This article belongs to the Topic Artificial Intelligence Applications in Financial Technology)

► Show Figures

Figure 1

16 pages, 7595 KiB

Open AccessArticle

Using Deep Reinforcement Learning with Hierarchical Risk Parity for Portfolio Optimization

by Adrian Millea and Abbas Edalat

Int. J. Financial Stud. 2023, 11(1), 10; https://doi.org/10.3390/ijfs11010010 - 29 Dec 2022

Cited by 7 | Viewed by 5415

Abstract

We devise a hierarchical decision-making architecture for portfolio optimization on multiple markets. At the highest level a Deep Reinforcement Learning (DRL) agent selects among a number of discrete actions, representing low-level agents. For the low-level agents, we use a set of Hierarchical Risk [...] Read more.

We devise a hierarchical decision-making architecture for portfolio optimization on multiple markets. At the highest level a Deep Reinforcement Learning (DRL) agent selects among a number of discrete actions, representing low-level agents. For the low-level agents, we use a set of Hierarchical Risk Parity (HRP) and Hierarchical Equal Risk Contribution (HERC) models with different hyperparameters, which all run in parallel, off-market (in a simulation). The information on which the DRL agent decides which of the low-level agents should act next is constituted by the stacking of the recent performances of all agents. Thus, the modelling resembles a statefull, non-stationary, multi-arm bandit, where the performance of the individual arms changes with time and is assumed to be dependent on the recent history. We perform experiments on the cryptocurrency market (117 assets), on the stock market (46 assets) and on the foreign exchange market (28 pairs) showing the excellent robustness and performance of the overall system. Moreover, we eliminate the need for retraining and are able to deal with large testing sets successfully. Full article

(This article belongs to the Topic Artificial Intelligence Applications in Financial Technology)

► Show Figures

Figure 1

15 pages, 252 KiB

Open AccessArticle

Financial Fraud Detection of Listed Companies in China: A Machine Learning Approach

by Yasheng Chen and Zhuojun Wu

Sustainability 2023, 15(1), 105; https://doi.org/10.3390/su15010105 - 21 Dec 2022

Cited by 7 | Viewed by 5398

Abstract

As the focus of capital market supervision, financial report fraud has shown a development trend of enormous numbers, complex transactions, and hidden means in recent years. To improve audit efficiency and reduce the dependence on non-financial data, the study only uses the structured [...] Read more.

As the focus of capital market supervision, financial report fraud has shown a development trend of enormous numbers, complex transactions, and hidden means in recent years. To improve audit efficiency and reduce the dependence on non-financial data, the study only uses the structured original data in the financial report to constructs a new fraud identification model, which can quickly detect fraud in China. This study takes the listed companies in China from 1998 to 2016 as research samples and selects 28 sets of raw data from financial reports. Then, this study compares the detection effectiveness of two single classification machine learning algorithms and five ensemble learning algorithms on fraud detection. Compared with single classification machine learning algorithms, the results show that ensemble learning algorithms are generally better at detecting fraud for Chinese listed companies, and the stacking algorithm performs the best. The study results provide direct evidence for rapid fraud detection using financial report raw data and ensemble learning algorithms. The study first proposes a stacking algorithm-based financial reporting fraud identification model for listed companies in China, which provides a simple and effective approach for investors, regulators, and management. It can also provide a reference for the detection of other fraud scenarios. Full article

(This article belongs to the Topic Artificial Intelligence Applications in Financial Technology)

17 pages, 466 KiB

Open AccessArticle

A Full Population Auditing Method Based on Machine Learning

by Yasheng Chen, Zhuojun Wu and Hui Yan

Sustainability 2022, 14(24), 17008; https://doi.org/10.3390/su142417008 - 19 Dec 2022

Cited by 1 | Viewed by 3055

Abstract

As it is urgent to change the traditional audit sampling method that is based on manpower to meet the growing audit demand in the era of big data. This study uses empirical methods to propose a full population auditing method based on machine [...] Read more.

As it is urgent to change the traditional audit sampling method that is based on manpower to meet the growing audit demand in the era of big data. This study uses empirical methods to propose a full population auditing method based on machine learning. This method can extend the application scope of the audit to all samples through the self-learning feature of machine learning, which helps to address the dependence on auditors’ personal experience and the audit risks arising from audit sampling. First, this paper demonstrates the feasibility of this method, then selects the financial data of a large enterprise for full population testing, and finally summarizes the critical steps of practical applications. The study results indicate that machine learning for full population auditing is able to detect, in all samples, abnormal business whose execution does not adhere to existing accounting rules, as well as abnormal business with irregular accounting rules, thus improving the efficiency of internal control audits. By combining the learning ability of machine-learning algorithms and the arithmetic power of computers, the proposed full population auditing method provides a feasible approach for the intellectual development of future auditing at the application level. Full article

(This article belongs to the Topic Artificial Intelligence Applications in Financial Technology)

► Show Figures

Figure 1

19 pages, 4372 KiB

Open AccessArticle

Digital Inclusive Finance and Family Wealth: Evidence from LightGBM Approach

by Ying Liu, Haoran Zhao, Jieguang Sun and Yahui Tang

Sustainability 2022, 14(22), 15363; https://doi.org/10.3390/su142215363 - 18 Nov 2022

Cited by 6 | Viewed by 2372

Abstract

With the rapid development of digital technology in China, Digital Inclusive Finance, which uses digital financial services to promote financial inclusion, is developing rapidly. This paper uses the Peking University Digital Financial Inclusion index of China and China Family Panel Studies (CFPS) data [...] Read more.

With the rapid development of digital technology in China, Digital Inclusive Finance, which uses digital financial services to promote financial inclusion, is developing rapidly. This paper uses the Peking University Digital Financial Inclusion index of China and China Family Panel Studies (CFPS) data to construct a predictive model using the LightGBM machine learning algorithm to study whether Digital Inclusive Finance can predict household wealth and analyze the characteristics of strong predictive ability for household wealth. They found that: (1) the introduction of the Digital Financial Inclusion index can improve the prediction performance of the household wealth model; (2) financial literacy and age characteristics are the key characteristics of household wealth accumulation; (3) the coverage and depth of Digital Inclusive Finance has a significant effect on family wealth accumulation, but the degree of digitization acts as a disincentive factor. This paper not only uses machine learning methods to do research on Digital Inclusive Finance and family wealth from a more comprehensive perspective, but also provides effective theoretical support for the key factors that enhance family wealth. Full article

(This article belongs to the Topic Artificial Intelligence Applications in Financial Technology)

► Show Figures

Figure 1

14 pages, 1154 KiB

Open AccessArticle

Machine Learning to Develop Credit Card Customer Churn Prediction

by Dana AL-Najjar, Nadia Al-Rousan and Hazem AL-Najjar

J. Theor. Appl. Electron. Commer. Res. 2022, 17(4), 1529-1542; https://doi.org/10.3390/jtaer17040077 - 16 Nov 2022

Cited by 27 | Viewed by 10228

Abstract

The credit card customer churn rate is the percentage of a bank’s customers that stop using that bank’s services. Hence, developing a prediction model to predict the expected status for the customers will generate an early alert for banks to change the service [...] Read more.

The credit card customer churn rate is the percentage of a bank’s customers that stop using that bank’s services. Hence, developing a prediction model to predict the expected status for the customers will generate an early alert for banks to change the service for that customer or to offer them new services. This paper aims to develop credit card customer churn prediction by using a feature-selection method and five machine learning models. To select the independent variables, three models were used, including selection of all independent variables, two-step clustering and k-nearest neighbor, and feature selection. In addition, five machine learning prediction models were selected, including the Bayesian network, the C5 tree, the chi-square automatic interaction detection (CHAID) tree, the classification and regression (CR) tree, and a neural network. The analysis showed that all the machine learning models could predict the credit card customer churn model. In addition, the results showed that the C5 tree machine learning model performed the best in comparison with the three developed models. The results indicated that the top three variables needed in the development of the C5 tree customer churn prediction model were the total transaction count, the total revolving balance on the credit card, and the change in the transaction count. Finally, the results revealed that merging the multi-categorical variables into one variable improved the performance of the prediction models. Full article

(This article belongs to the Topic Artificial Intelligence Applications in Financial Technology)

► Show Figures

Figure 1

Topic Menu

Topic Editors

Artificial Intelligence Applications in Financial Technology

Topic Information

Keywords

Participating Journals

Published Papers (20 papers)

Further Information

Guidelines

MDPI Initiatives

Follow MDPI