PERFORMA LARGE LANGUAGE MODEL DALAM OTOMASI PEMBUATAN JURNAL DAN REKONSILIASI KEUANGAN: SIMULASI PADA SISTEM INFORMASI AKUNTANSI BERBASIS CLOUD
PERFORMANCE OF LARGE LANGUAGE MODELS IN AUTOMATING JOURNAL ENTRY GENERATION AND FINANCIAL RECONCILIATION: A SIMULATION IN A CLOUD-BASED ACCOUNTING INFORMATION SYSTEM
DOI:
https://doi.org/10.52972/hoaq.vol17no2.p292-305Keywords:
Large Language Model (LLM), otomasi pembuatan jurnal, rekonsiliasi keuangan, sistem informasi akuntansi berbasis cloud, Automated Journal Entry GenerationAbstract
Transformasi digital dalam sistem informasi akuntansi membuka peluang pemanfaatan Large Language Model (LLM) untuk mengotomasi pembuatan jurnal dan rekonsiliasi keuangan, terutama pada KUD/UMKM pengolahan sawit yang masih menghadapi keterbatasan pencatatan manual dan semi-digital. Penelitian ini bertujuan mengevaluasi akurasi, kecepatan, dan keandalan beberapa model LLM dalam menghasilkan jurnal otomatis dan melakukan rekonsiliasi transaksi pada Sistem Informasi Akuntansi berbasis cloud, serta mengidentifikasi pengaruh kompleksitas transaksi dan kualitas data terhadap performanya. Penelitian ini menggunakan pendekatan kuantitatif eksperimental berbasis simulasi dengan desain faktorial yang membandingkan GPT-4, Llama 3, Mistral, dan model FinEdge-like pada dataset sintetis transaksi dan paket rekonsiliasi yang dikonstruksi berdasarkan karakteristik operasional KUD/UMKM pengolahan sawit. Evaluasi dilakukan menggunakan metrik klasifikasi, waktu pemrosesan, error rate, statistik deskriptif, dan ANOVA tiga-arah. Berdasarkan eksperimen terhadap dataset sintetis yang dikonstruksi dan disimulasikan sesuai desain penelitian, GPT-4 menunjukkan performa tertinggi, diikuti FinEdge-like, Llama 3, dan Mistral. Seluruh model mengalami penurunan akurasi ketika kompleksitas transaksi meningkat dan kualitas data menurun. LLM juga menunjukkan waktu pemrosesan yang jauh lebih singkat dibandingkan prosedur manual dalam skenario simulasi. Temuan ini menegaskan potensi integrasi LLM dalam SIA berbasis cloud sebagai mekanisme augmentasi akuntansi yang tetap memerlukan validasi manusia, audit trail, dan pengendalian berbasis risiko. Temuan ini merepresentasikan kinerja model dalam skenario simulasi dan belum dapat ditafsirkan sebagai bukti kinerja pada data transaksi lapangan aktual.
Digital transformation in accounting information systems opens up opportunities to utilize Large Language Models (LLMs) to automate journal entry generation and financial reconciliation, particularly for palm oil processing cooperatives and MSMEs that still face limitations with manual and semi-digital record-keeping. This study aims to evaluate the accuracy, speed, and reliability of several LLM models in generating automated journal entries and performing transaction reconciliations in a cloud-based accounting information system, as well as to identify the impact of transaction complexity and data quality on their performance. This study employs a quantitative experimental approach based on simulation with a factorial design that compares GPT-4, Llama 3, Mistral, and FinEdge-like models on a synthetic dataset of transactions and reconciliation packages constructed based on the operational characteristics of palm oil processing KUDs/SMEs. Evaluation was conducted using classification metrics, processing time, error rate, descriptive statistics, and a three-way ANOVA. Based on experiments using the synthetic dataset constructed and simulated according to the study design, GPT-4 demonstrated the highest performance, followed by the FinEdge-like model, Llama 3, and Mistral. All models experienced a decline in accuracy as transaction complexity increased and data quality decreased. LLMs also demonstrated significantly shorter processing times compared to manual procedures in the simulation scenarios. These findings underscore the potential for integrating LLMs into cloud-based SIA as an accounting augmentation mechanism that still requires human validation, audit trails, and risk-based controls. These findings represent model performance in simulation scenarios and cannot yet be interpreted as evidence of performance on actual field transaction data.
References
C. Sampaio and R. Silva, “Digital Transformation in Accounting: An Assessment of Automation and AI Integration,” International Journal of Financial Studies, vol. 13, no. 4, p. 206, Nov. 2025, doi: 10.3390/ijfs13040206.
Y. Nie et al., “A Survey of Large Language Models for Financial Applications: Progress, Prospects and Challenges,” Jun. 2024.
S. Oviya, N. Sharadha, T. Bhuvaneswari, S. Vijayalakshmi, and S. Sushma, “The Impact of Automation and Ai in Revolutionising Traditional Accounting Methods,” Journal of Informatics Education and Research, 2024, doi: 10.52783/jier.v4i2.1001.
F. R. P. Faiz, S. A. D. Safina, C. A. Citra, A. Angelia, I. N. Iqlima, and C. D. H. N. Christyan, “Systematic Literatur Review: Pengaruh Cloud Accounting Terhadap Efektivitas Sistem Pelaporan Keuangan Di UMKM,” Jurnal Ekonomika Dan Bisnis (JEBS), vol. 5, no. 3, pp. 527–533, May 2025, doi: 10.47233/jebs.v5i3.2769.
J. Gupta et al., “Large Language Models Acing Chartered Accountancy,” SN Comput. Sci., vol. 6, no. 8, p. 957, Nov. 2025, doi: 10.1007/s42979-025-04497-x.
S. Liu et al., “FinDABench: Benchmarking Financial Data Analysis Ability of Large Language Models,” Jun. 2024.
K. Akshat, “AI-Driven Reconciliation Agents for Financial Accuracy and Compliance in Cloud-Native Data Pipelines,” International Research Journal on Advanced Engineering Hub (IRJAEH), vol. 3, no. 09, pp. 3502–3508, Sep. 2025, doi: 10.47392/IRJAEH.2025.0514.
D. Permatasari, N. F. Mohammed, N. A. Shafie, and M.-E. Sánchez-del-Río-Vázquez, “Technological-Organisational-Environmental Determinants of Cloud Accounting System Usage in Indonesian SMEs,” Journal of Nusantara Studies (JONUS), vol. 10, no. 2, pp. 566–589, Jul. 2025, doi: 10.24200/jonus.vol10iss2pp566-589.
A. M. Musyaffi et al., “Exploring Technological Factors and Cloud Accounting Adoption in MSMEs: A Comprehensive TAM Framework,” International Review of Management and Marketing, vol. 15, no. 1, pp. 283–292, Dec. 2024, doi: 10.32479/irmm.17451.
J. Slezák, “Modeling Technology Acceptance in Accounting Area,” Engineering Economics, vol. 36, no. 3, pp. 291–302, Jun. 2025, doi: 10.5755/j01.ee.36.3.33645.
D. Permatasari, N. Farizal, and N. Aima, “Exploring Factors Influencing the Adoption of Cloud Accounting Systems in Indonesian Micro Small and Medium Enterprises: A Unified Theory of Acceptance and Use of Technology Based Analysis,” Management and Accounting Review, vol. 23, no. 1, Apr. 2024, doi: 10.24191/MAR.V23i01-07.
G. Bhatia, E. M. B. Nagoudi, H. Cavusoglu, and M. Abdul-Mageed, “FinTral: A Family of GPT-4 Level Multimodal Financial Large Language Models,” Jun. 2024.
N. Vali, S. Saha, S. Koora, and Md. A. Vali, “Automating Financial Workflows: AI-Powered Revenue Reconciliation for Real Estate Enterprises,” International Journal of Financial Management and Economics, vol. 8, no. 2, pp. 273–283, Jan. 2025, doi: 10.33545/26179210.2025.v8.i2.595.
D. Qoriah, “Adoption of Cloud Accounting: Opportunities, Barriers, and Implications for SMEs,” Summa?: Journal of Accounting and Tax, vol. 3, no. 3, pp. 191–204, Jul. 2025, doi: 10.61978/summa.v3i3.979.
G. Matlin, M. Okamoto, H. Pardawala, Y. Yang, and S. Chava, “Financial Language Model Evaluation (FLaME),” in Findings of the Association for Computational Linguistics: ACL 2025, Stroudsburg, PA, USA: Association for Computational Linguistics, 2025, pp. 22633–22679. doi: 10.18653/v1/2025.findings-acl.1164.
R. S. Pall, “IndiaFinBench: An Evaluation Benchmark for Large Language Model Performance on Indian Financial Regulatory Text,” May 2026.
M. Taqi, “AI Application in Accounting Studies,” Review on Islamic Accounting, vol. 4, no. 2, Feb. 2025, doi: 10.58968/ria.v4i2.598.
Z. Li and H. Qin, “Large Language Models (LLMs) in Business Strategies and Accounting: Opportunities and Challenges,” Management Journal for Advanced Research, vol. 4, no. 6, pp. 1–7, Dec. 2024, doi: 10.54741/mjar.4.6.1-7.
A. P. Oktavia, A. Nuraliati, and L. Ulfiana, “The Role of Cloud Accounting in Improving MSME Performance in The Digital Economy Era,” Oikonomia?: Journal of Management Economics and Accounting, vol. 3, no. 1, pp. 1–10, Dec. 2025, doi: 10.61942/oikonomia.v3i1.494.
P. H. Mainardi and E. D. Bidoia, “Fundamental Concepts And Recent Applications of Factorial Statistical Designs,” Brazilian Journal of Biometrics, vol. 40, no. 1, Mar. 2022, doi: 10.28951/bjb.v40i1.552.
R. S. Pall, “IndiaFinBench: An Evaluation Benchmark for Large Language Model Performance on Indian Financial Regulatory Text,” May 2026.
S. Y. A. Safavi-Naini et al., “Benchmarking Proprietary and Open-Source Language and Vision-Language Models for Gastroenterology Clinical Reasoning,” NPJ Digit. Med., vol. 8, no. 1, p. 797, Nov. 2025, doi: 10.1038/s41746-025-02174-0.
T.-T.-H. Le, A. A. Adiputra, J. Yun, and H. Kim, “Anomaly Detection in Industrial Machine Sounds Using High-Frequency Features and Gate Recurrent Unit Networks,” IEEE Access, vol. 13, pp. 77165–77186, 2025, doi: 10.1109/ACCESS.2025.3565812.
X. Peng et al., “Plutus: Benchmarking Large Language Models in Low-Resource Greek Finance,” Feb. 2025.
S. Noels, J. De Blaere, and T. De Bie, “A Dutch Financial Large Language Model,” in Proceedings of the 5th ACM International Conference on AI in Finance, New York, NY, USA: ACM, Nov. 2024, pp. 283–291. doi: 10.1145/3677052.3698628.
M. M. Bishal, Md. R. H. Chowdory, A. Das, and M. A. Kabir, “COVIDHealth: A novel Labeled Dataset and Machine Learning-Based Web Application for Classifying COVID-19 discourses on Twitter,” Heliyon, vol. 10, no. 14, p. e34103, Jul. 2024, doi: 10.1016/j.heliyon.2024.e34103.
T. J. Nasser and B. K. Mohammed, “Estimate Factorial Experimental Design by Using Some Methods With an Application,” International Journal of Statistics and Applied Mathematics, vol. 10, no. 8S, pp. 43–52, Aug. 2025, doi: 10.22271/maths.2025.v10.i8Sa.2149.
R. Ramírez-Tapia, A. J. Ríos-Lira, Y. V. Pantoja-Pacheco, J. A. Vázquez-López, and E. A. Ruelas-Santoyo, “Quantitative Analysis of the Balance Property in Factorial Experimental Designs 24 to 28,” Mathematics, vol. 10, no. 20, p. 3812, Oct. 2022, doi: 10.3390/math10203812.
D. Stanishevskii et al., “FINESSE-Bench: A Hierarchical Benchmark Suite for Financial Domain Knowledge and Technical Analysis in Large Language Models,” May 2026.
C. Yang, C. Xu, and Y. Qi, “Financial Knowledge Large Language Model,” Jun. 2024.
Y. Nie et al., “A Survey of Large Language Models for Financial Applications: Progress, Prospects and Challenges,” Jun. 2024.
A. Jadhav and V. Mirza, “Large Language Models in Equity Markets: Applications, Techniques, and Insights,” Front. Artif. Intell., vol. 8, Aug. 2025, doi: 10.3389/frai.2025.1608365.
A. Srivastava, S. Jung, and S. Mateega, “Market-Bench: Evaluating Large Language Models on Introductory Quantitative Trading and Market Dynamics,” Jan. 2026.
X. Peng et al., “Ebisu: Benchmarking Large Language Models in Japanese Finance,” Feb. 2026.
M. Birti, A. Maurino, and F. Osborne, “Optimizing Large Language Models for ESG Activity Detection in Financial Texts,” in Proceedings of the 6th ACM International Conference on AI in Finance, New York, NY, USA: ACM, Nov. 2025, pp. 856–863. doi: 10.1145/3768292.3770371.
D. Shu, H. Yuan, Y. Wang, Y. Liu, H. Zhang, and M. Du, “FinChart-Bench: Benchmarking Financial Chart Comprehension in Vision-Language Models,” in Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Stroudsburg, PA, USA: Association for Computational Linguistics, 2026, pp. 13447–13466. doi: 10.18653/v1/2026.acl-long.615.
K. N. Sumiari, I. K. Parnata, I. G. A. A. Pramitari, and I. M. A. Putrayasa, “The Emergence of Large Language Models in Financial Report Auditing: Opportunities, Benchmarks, Risks, and The Road Ahead,” Greenation International Journal of Economics and Accounting, vol. 4, no. 2, pp. 364–373, Jun. 2026, doi: 10.38035/gijea.v4i2.901.
Z. Liu, X. Zhang, K. Yang, Q. Xie, J. Huang, and S. Ananiadou, “FMDLlama: Financial Misinformation Detection Based on Large Language Models,” in Companion Proceedings of the ACM on Web Conference 2025, New York, NY, USA: ACM, May 2025, pp. 1153–1157. doi: 10.1145/3701716.3715599.
D. Lee, H. Yun, M. Cha, S. Park, S. Park, and J. Kim, “EconCausal: A Context-Aware Economic Reasoning Benchmark for Large Language Models,” May 2026.
S. Lin et al., “Evaluation and Benchmarking Suite for Financial Large Language Models and Agents,” Feb. 2026.
D. Chandra, “Applications of Large Language Model Reasoning in Feature Generation,” Mar. 2025.
Y. He et al., “Are Large Language Models Reliable Reviewers? A Benchmark for Error Detection in Financial Documents,” in Findings of the Association for Computational Linguistics: ACL 2026, Stroudsburg, PA, USA: Association for Computational Linguistics, 2026, pp. 29625–29643. doi: 10.18653/v1/2026.findings-acl.1481.
C. Woesle, L. Fischer-Brandies, and R. Buettner, “A Systematic Literature Review of Hallucinations in Large Language Models,” IEEE Access, vol. 13, pp. 148231–148253, 2025, doi: 10.1109/ACCESS.2025.3601206.
Z. Xie et al., “The CLEF-2026 FinMMEval Lab: Multilingual and Multimodal Evaluation of Financial AI Systems,” Feb. 2026.
K. Veeramani, A. Joseph N, and P. M, “Neuro-Symbolic Reasoning Engine for Tax Optimisation,” Front. Artif. Intell., vol. 9, May 2026, doi: 10.3389/frai.2026.1802755.
A. Hamzah, D. Suhendar, and A. Z. Arifin, “Factors Affecting Cloud Accounting Adoption In SMEs,” Jurnal Akuntansi, vol. 27, no. 3, pp. 442–464, Sep. 2023, doi: 10.24912/ja.v27i3.1520.
E. Novianti and F. Aligarh, “MSME Adoption of Cloud Accounting: A TAM and UTAUT Perspective,” Proceedings of Economics Business Innovation & Creativity, vol. 2, no. 1, pp. 442–463, Jun. 2025, doi: 10.32493/ebic.v2i1.51050.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 HOAQ (High Education of Organization Archive Quality) : Jurnal Teknologi Informasi
This work is licensed under a Creative Commons Attribution 4.0 International License.
Jurnal HOAQ (High Education of Organization Archive Quality) : Jurnal Teknologi Informasi diterbitkan berdasarkan lisensi Creative Commons Attribution 4.0 International License (CC BY 4.0). Lisensi ini memungkinkan setiap orang untuk Berbagi: menyalin dan mendistribusikan kembali materi ini dalam format atau bentuk apapun; Adaptasi: merombak, mengubah, dan membuat turunan dari materi ini untuk keperluan apa pun, termasuk keperluan komersial, asalkan mereka memberikan pengakuan kepada Penulis Asli atas hasil karya aslinya.
