Deep learning enhancing banking services: a hybrid transaction classification and cash flow prediction approach

Kotios, Dimitrios; Makridis, Georgios; Fatouros, Georgios; Kyriazis, Dimosthenis

doi:10.1186/s40537-022-00651-x

Cited by 19 publications

(9 citation statements)

References 33 publications

(42 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The non NLP-based SVM solution for cash flow prediction for small & medium enterprises by D. Kotios et al (2022) [31] attained a precision that was only 0.2 % higher than ours, after trying many other algorithms.…”

Section: E Comparison With Prior Workcontrasting

confidence: 45%

“…• The main contribution of this study is a novel solution for automatic industrial CF estimation from bank transactions based on supervised ML and NLP techniques. • The performance of the underlying bank transaction classification methodology is comparable to that of other researchers [30], [31]. • An experimental data-set composed of more than 25,000 bank transactions was used.…”

Section: Discussionmentioning

confidence: 77%

“…The agnostic explanation methodology extracts a set of relevant words for the classifier, and this explanation set is then validated with a similarity metric by comparing the 31 Available at https://es.dbpedia.org, November 2022. 32 Available at https://www.wikidata.org, November 2022.…”

Section: Discussionmentioning

confidence: 99%

“…Nevertheless, no previous work on bank transaction classification had an XAI perspective (with the sole exception of Kotios el al. (2022) [31], although it did not involve any NLP methodology) nor considered industrial CF estimation.…”

Section: Related Workmentioning

confidence: 99%

See 3 more Smart Citations

Explainable Automatic Industrial Carbon Footprint Estimation From Bank Transaction Classification Using Natural Language Processing

et al. 2022

View full text Add to dashboard Cite

Concerns about the effect of greenhouse gases have motivated the development of certification protocols to quantify the industrial carbon footprint (CF). These protocols are manual, workintensive, and expensive. All of the above have led to a shift towards automatic data-driven approaches to estimate the CF, including Machine Learning (ML) solutions. Unfortunately, as in other sectors of interest, the decision-making processes involved in these solutions lack transparency from the end user's point of view, who must blindly trust their outcomes compared to intelligible traditional manual approaches. In this research, manual and automatic methodologies for CF estimation were reviewed, taking into account their transparency limitations. This analysis led to the proposal of a new explainable ML solution for automatic CF calculations through bank transaction classification. Consideration should be given to the fact that no previous research has considered the explainability of bank transaction classification for this purpose. For classification, different ML models have been employed based on their promising performance in similar problems in the literature, such as Support Vector Machine, Random Forest, and Recursive Neural Networks. The results obtained were in the 90 % range for accuracy, precision, and recall evaluation metrics. From their decision paths, the proposed solution estimates the CO 2 emissions associated with bank transactions. The explainability methodology is based on an agnostic evaluation of the influence of the input terms extracted from the descriptions of transactions using locally interpretable models. The explainability terms were automatically validated using a similarity metric over the descriptions of the target categories. Conclusively, the explanation performance is satisfactory in terms of the proximity of the explanations to the associated activity sector descriptions, endorsing the trustworthiness of the process for a human operator and end users.

show abstract

Section: E Comparison With Prior Workcontrasting

confidence: 45%

Section: Discussionmentioning

confidence: 77%

Section: Discussionmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Explainable Automatic Industrial Carbon Footprint Estimation From Bank Transaction Classification Using Natural Language Processing

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Globally, small and medium sized enterprises (SMEs) make up around 99% of the world's businesses and they employ 50-60% of the world's population (Kotios, Makridis, Fatouros & Kyriazis, 2022). In Europe, SMEs account for up to 99.8% of all its firms.…”

Section: Introductionmentioning

confidence: 99%

Effect of Cash Management Practices on Financial Performance of Small and Medium Enterprises in Kenya. A Case Study of Mombasa County

RICHO,

ONYIEGO, PhD

2023

strategicjournals.com

View full text Add to dashboard Cite

This study examined the effect of cash management practices on the financial performance of small and medium enterprises in Kenya. The study was anchored on four theories: The Rational Choice Theory, Modern portfolio theory, the cash flow theory and the cash management theory. A survey research design was adopted in the study. The researcher used census to select 98 respondents from all the registered hardware operating in Mombasa County. The researcher personally called the accountants and read for them the questionnaires and recorded the data. The data was tabulated, and then analyzed using descriptive and inferential statistics with the help of Statistical package for the Social Sciences (SPSS) version 26 software. Quantitative data was analyzed using descriptive statistics which included frequencies, percentages, mean and standard deviation. Inferential statistics included correlations and regression analysis using the model Y = B0 + B1X1+ B2X2+B3X3 + B4X4 + e to test the effect of the independent on the dependent variable. The results of the study were presented using tables. The study found that there is a significantly positive effect of bank reconciliation on the financial performance of SMEs in Kenya. The study also established that there is a significantly positive effect of credit management on the financial performance of SMEs in Kenya. The study found that cash planning had a positively significant effect on the financial performance of SMEs in Kenya. The study also established that cash flow had a significantly positive effect on that financial performance of SMEs in Kenya. The study concluded that cash management practices had a significantly positive effect on the financial performance of SMEs in Kenya. The study recommended that cash management practices such as bank reconciliation, credit management, cash planning and cash flow should be adopted by SMEs in Kenya in order to enhance their financial performance

show abstract

Evaluating end‐users' digital competencies and ethical perceptions of AI systems in the context of sustainable digital banking

Fundira,

Edoun,

Pradhan

2024

Sustainable Development

View full text Add to dashboard Cite

In the pursuit of fostering sustainable digital banking (SDB), this study investigates the nexus between end‐users' digital competencies, perceptions of artificially intelligent systems, and sustainable practices. With a focus on addressing inclusivity barriers and promoting ethical considerations, expert interviews were employed to provide nuanced insights. The study's foundation rests on the significance of sustainable development goals, digital competencies, and artificial intelligence (AI) ethics in shaping the future of digital banking. An interview schedule was developed and distributed online to IT experts and researchers in the field of AI. The instrument examined the major challenges faced by customers in accessing digital banking, the expected timeline for the complete digitalization of the South African banking sector, customers' anticipated competency, and comfort levels with a 100% digital banking interface, and their perspectives on the overall impact of such a transition on humanity. The interview gathered participants' views on AI‐based banking systems that determine credit scores and creditworthiness. The study also included questions to obtain expert opinions on customers' digital capabilities, access to resources, and the impact of AI trust issues on financial service customers. The findings of this study provided valuable insights for stakeholders, including governments, NGOs, and financial institutions. Implications derived from expert perspectives illuminate strategies for enhancing digital competencies, addressing inclusivity hurdles, and ensuring ethical AI integration. This study contributes to the sustainable development discourse by offering actionable recommendations for stakeholders in the pursuit of a more inclusive, digitally competent, and ethically responsible digital banking landscape.

show abstract

Deep learning enhancing banking services: a hybrid transaction classification and cash flow prediction approach

Cited by 19 publications

References 33 publications

Explainable Automatic Industrial Carbon Footprint Estimation From Bank Transaction Classification Using Natural Language Processing

Explainable Automatic Industrial Carbon Footprint Estimation From Bank Transaction Classification Using Natural Language Processing

Effect of Cash Management Practices on Financial Performance of Small and Medium Enterprises in Kenya. A Case Study of Mombasa County

Evaluating end‐users' digital competencies and ethical perceptions of AI systems in the context of sustainable digital banking

Contact Info

Product

Resources

About