Integrate AI-based chatbots into accounting services: Enhance customer communication and financial management support DOI
Ziqi He,

Xue Jin

Journal of Computational Methods in Sciences and Engineering, Journal Year: 2025, Volume and Issue: unknown

Published: April 30, 2025

As businesses increasingly adopt digital tools to streamline operations, artificial intelligence (AI)-based chatbots have emerged as vital components for enhancing customer communication and supporting financial management within accounting services. This research focuses on reliable AI-powered capable of handling complex tasks while user satisfaction. The goal this study is establish AI-based in services improve assistance communication. paper presents a novel Raven Roosting-tuned Adaptive Bidirectional Long Short-Term Memory (RR-ABiLSTM) model designed classify queries enhance contextual understanding conversations communications. dataset encompasses both structured unstructured data from conversations, constituting domain-specific corpus focusing common tasks. Data preprocessing included text cleaning tokenization applied the acquired data. Subsequently, feature extraction was performed using Word2Vec. RR algorithm utilized optimize hyperparameters selection, BiLSTM ensures deep relationships thereby accuracy efficiency processing queries. Furthermore, dynamic training mechanism integrated, allowing chatbot continually adapt increasing consumer demands without downtime. proposed method implemented Python software, its performance compared with traditional algorithms. overall metrics—F1-score (87.75%), precision (89.25%), recall (86.24%), (90%)—illustrate that suggested significantly improves engagement, reduces workload accountants, enhances by providing support.

Language: Английский

AI in Banking: What Drives Generation Z to Adopt AI-Enabled Voice Assistants in Saudi Arabia? DOI Creative Commons
Rotana S. Alkadi, Salma S. Abed

International Journal of Financial Studies, Journal Year: 2025, Volume and Issue: 13(1), P. 36 - 36

Published: March 3, 2025

The aim of this study is to examine the factors that drive Saudi Arabian Generation Z’s intention use voice assistants (VAs) in banking. Technology Acceptance Model (TAM) was extended by incorporating three additional constructs: subjective norms, which capture social influence close relationships, including family and friends; personal innovativeness, reflects openness new technologies characteristic Z; perceived trust, addresses concerns related security reliability are critical financial contexts, thereby enhancing our understanding phenomenon among Z. A survey 292 Z respondents collected structural equation modeling (SEM) employed for data analysis. findings reveal such as usefulness, attitude, trust all have a significantly positive impact on AI-enabled VAs Additionally, results indicate usefulness influenced ease use, while attitude affected trust. Despite government’s support initiatives development AI-fintech industry, there still lack about consumer behavioral toward Arabia and, particularly This contributes existing literature provides valuable recommendations policymakers fintech service providers seeking implement effective enrich consumers’ engagement experience.

Language: Английский

Citations

0
Integrate AI-based chatbots into accounting services: Enhance customer communication and financial management support DOI
Ziqi He,

Xue Jin

Journal of Computational Methods in Sciences and Engineering, Journal Year: 2025, Volume and Issue: unknown

Published: April 30, 2025

As businesses increasingly adopt digital tools to streamline operations, artificial intelligence (AI)-based chatbots have emerged as vital components for enhancing customer communication and supporting financial management within accounting services. This research focuses on reliable AI-powered capable of handling complex tasks while user satisfaction. The goal this study is establish AI-based in services improve assistance communication. paper presents a novel Raven Roosting-tuned Adaptive Bidirectional Long Short-Term Memory (RR-ABiLSTM) model designed classify queries enhance contextual understanding conversations communications. dataset encompasses both structured unstructured data from conversations, constituting domain-specific corpus focusing common tasks. Data preprocessing included text cleaning tokenization applied the acquired data. Subsequently, feature extraction was performed using Word2Vec. RR algorithm utilized optimize hyperparameters selection, BiLSTM ensures deep relationships thereby accuracy efficiency processing queries. Furthermore, dynamic training mechanism integrated, allowing chatbot continually adapt increasing consumer demands without downtime. proposed method implemented Python software, its performance compared with traditional algorithms. overall metrics—F1-score (87.75%), precision (89.25%), recall (86.24%), (90%)—illustrate that suggested significantly improves engagement, reduces workload accountants, enhances by providing support.

Language: Английский

Citations

0