Research on Small-Sample Credit Card Fraud Identification Based on Temporal Attention-Boundary-Enhanced Prototype Network DOI Creative Commons
Boyu Liu,

Licheng Wu,

Shengdong Mu

и другие.

Mathematics, Год журнала: 2024, Номер 12(24), С. 3894 - 3894

Опубликована: Дек. 10, 2024

The Nielsen Report points out that credit card fraud caused business losses of USD 28.65 billion globally in 2019, with the US accounting for more than one-third high share, and insufficient identification has brought about a serious loss financial institutions’ ability to do business. In small sample data environments, traditional detection methods based on prototype network models struggle time-series features challenge identifying uncorrected distribution metric space. this paper, we propose method called Time-Series Attention-Boundary-Enhanced Prototype Network (TABEP), which strengthens temporal feature dependency between channels by incorporating attention module achieve channel fusion acquisition. Additionally, nearest-neighbor boundary is introduced after computation prototype-like model adjust overall space clarify representation boundaries model. Experimental results show TABEP achieves higher accuracy compared five existing baseline methods, better fits distribution, significantly improves performance. This study highlights effectiveness open innovation addressing complex security problems, great significance promoting technological advancement field security.

Язык: Английский

Factors affecting chatbots in banking services: the UTAUT2 and innovation resistance theory perspective DOI Creative Commons
Zericho R. Marak, Subhajit Pahari, Rishabh Shekhar

и другие.

Journal of Innovation and Entrepreneurship, Год журнала: 2025, Номер 14(1)

Опубликована: Май 6, 2025

Язык: Английский

Процитировано

0

EVALUATING TECHNOLOGY ADOPTION RELEVANCE IN AUDIT AND NON-AUDIT FIRMS: A COMPARATIVE ANALYSIS DOI Open Access
Adriana Shamsudin, Mohd Noor Azli Ali Khan, Ahmad Jusoh

и другие.

Journal of Information System and Technology Management, Год журнала: 2024, Номер 9(35), С. 75 - 97

Опубликована: Июнь 30, 2024

Technology has substantially changed the accounting practice, impacting how financial information is processed, analysed, and reported. AI, RPA, Big Data Analytics are not used very often, which could be a sign of problems that make them hard to adopt. The rankings for relevance utilisation demonstrate practitioners see things rather than people actually use them. It necessary understand level relevancy various types technologies in practices. study aims determine current technology practices further outline each specifically employed both audit non-audit firms. quantitative nature approached ten (10) practitioners, consist executives, managers, senior management who directly responsible process preparing statements. participants were requested rank 17 established by Malaysian Institute Accountants (MIA). findings ranked assigned weightage item based on three (3) approaches. most relevant among firms with scores over 90% Microsoft Applications, Communication Technology, Mobile Fintech, Application Software, Online Communication. However, emerging like Artificial Intelligence, XBRL, Robotic Process Automation, remain relatively low, falling below 60%. If Weighted Cooke’s Index approach applied, blockchain will highest other due its rarity significant potential benefit profession. limited number exclusive focus specific professionals within field might limit applicability findings. Future research should investigate actual usage these broad scale validate results have been gathered comprehend factors impact importance real-world

Язык: Английский

Процитировано

2

E-invoicing: A Catalyst for Digitalization and Sustainability DOI Creative Commons

Rok Bojanc,

Andreja Pucihar, Gregor Lenart

и другие.

Organizacija, Год журнала: 2024, Номер 57(1), С. 3 - 19

Опубликована: Фев. 1, 2024

Abstract Background and Purpose Digitalization brings a wide range of opportunities for both digital sustainable transformation. One the first steps on this journey is digitization business documents digitalization processes. The European Commission has recognised advantages digitalization, particularly in context e-invoicing, which can contribute significantly to economic prosperity align with public policy goals such as deficit reduction development. Despite successful adoption e-invoicing sector, uptake business-to-business (B2B) transactions surprisingly low. This research focuses Slovenia, small country that imple-mented mandatory between enterprises sector 2015. Methods To investigate wider population, we designed an online survey, was conducted among 284 organizations Slovenia. collected data were analyzed using SPSS 28 software. sample t-test used evaluate importance identified advantages, barriers incentives usage. Further, ran inde-pendent identify statistically important differences ad-vantages, barriers, lower higher levels Results results showed recognize benefits well perceive them important. most are related environment, unawareness lack knowledge about how implement e-invoicing. Desired easier costs technical implementation, provided training education, including best practices, availability government grants other support measures state level, legislation, directory entities electronic document exchange service providers. Conclusion paper provides information decision-makers ad-min-istrations take appropriate further e-documents exploit better.

Язык: Английский

Процитировано

1

Selection and Adoption of Water Purification Technologies in the Bottled Water Industry in India DOI Creative Commons
Aviram Sharma, Saradindu Bhaduri

Cleaner Water, Год журнала: 2024, Номер unknown, С. 100039 - 100039

Опубликована: Сен. 1, 2024

Язык: Английский

Процитировано

1

Research on Small-Sample Credit Card Fraud Identification Based on Temporal Attention-Boundary-Enhanced Prototype Network DOI Creative Commons
Boyu Liu,

Licheng Wu,

Shengdong Mu

и другие.

Mathematics, Год журнала: 2024, Номер 12(24), С. 3894 - 3894

Опубликована: Дек. 10, 2024

The Nielsen Report points out that credit card fraud caused business losses of USD 28.65 billion globally in 2019, with the US accounting for more than one-third high share, and insufficient identification has brought about a serious loss financial institutions’ ability to do business. In small sample data environments, traditional detection methods based on prototype network models struggle time-series features challenge identifying uncorrected distribution metric space. this paper, we propose method called Time-Series Attention-Boundary-Enhanced Prototype Network (TABEP), which strengthens temporal feature dependency between channels by incorporating attention module achieve channel fusion acquisition. Additionally, nearest-neighbor boundary is introduced after computation prototype-like model adjust overall space clarify representation boundaries model. Experimental results show TABEP achieves higher accuracy compared five existing baseline methods, better fits distribution, significantly improves performance. This study highlights effectiveness open innovation addressing complex security problems, great significance promoting technological advancement field security.

Язык: Английский

Процитировано

1