Analysis of Data Quality and Reliability for AI-assisted Product Decision-Making from the Perspective of Cross-Platform User Analysis

Authors

  • Linhan Peng School of Computer Science, Faculty of Engineering, The University of Sydney, Sydney, NSW 2006, Australia

DOI:

https://doi.org/10.54097/n7rg5553

Keywords:

Cross-platform user analysis, AI-assisted decision-making, Data quality, Data reliability, Product decision-making

Abstract

The user data on different platforms are not uniform and therefore less reliable for AI-assisted product selection. Based on the differences in technical structure, user groups and data-collection methods, this paper examines how problems in data quality spread to the AI decision-making process. User interactions on different devices and browsers are often identified by different IDs. Failure of identity matching results in a loss of data integrity but improves the coverage of integrated data. The quality of the data on all platforms is inconsistent; only about half of the respondents have valid system version information in their user agent strings. Data readiness is the connection between AI capability and decision quality, and has an indirect effect of 0.476. There are also errors at the algorithm level. Even if the number of user segments is changed, and the data and algorithm are the same, the gender average deviation will be 54.5%. It can be seen that only by means of data cleaning or model correction is the reliability of decision-making guaranteed. Establish an uninterrupted translation path between business semantics and data semantics, explore how to guarantee the chain of data governance and decision calibration, and focus on studying the function of the calibration mechanism in human-machine collaboration.

Downloads

Download data is not yet available.

References

[1] Winter, M., Probst, T., Keil, T., & Pryss, R. (2025). A comparison of self-reported COVID-19 symptoms between android and iOS CoronaCheck app users. npj Digital Medicine, 8(1), 197. DOI: https://doi.org/10.1038/s41746-025-01595-1

[2] Lin, T., & Misra, S. (2022). Frontiers: The identity fragmentation bias. Marketing Science, 41(3), 433–440. DOI: https://doi.org/10.1287/mksc.2022.1360

[3] Powała, A., & Slezak, D. (2024). QEDrants–Data Quality Quadrants for Business Users and Decision-Makers. In Proceedings of FedCSIS (pp. 79–88). DOI: https://doi.org/10.15439/2024F0005

[4] Sambasivan, N., Kapania, S., Highfill, H., Akrong, D., Paritosh, P., & Aroyo, L. M. (2021, May). "Everyone wants to do the model work, not the data work": Data Cascades in High-Stakes AI. In Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems (pp. 1–15). DOI: https://doi.org/10.1145/3411764.3445518

[5] Airfocus by Lucid. (2026, May 26). New research: 500 product leaders reveal the biggest blockers to scaling AI. https://airfocus.com/blog/ai-product-management-scaling-blockers/

[6] Alexakis, T., Adamopoulou, E., Peppes, N., Daskalakis, E., & Ntouskas, G. (2025). Evaluating data quality: Comparative insights on standards, methodologies, and modern software tools. Electronics, 14(15), 3038. DOI: https://doi.org/10.3390/electronics14153038

[7] Hagn, M., Heinrich, B., Krapf, T., & Schiller, A. (2025). Handling imperfection: A taxonomy for machine learning on data with data quality defects. Decision Support Systems, 196, 114493. DOI: https://doi.org/10.1016/j.dss.2025.114493

[8] Laurito, W., Davis, B., Grietzer, P., Gavenčiak, T., Böhm, A., & Kulveit, J. (2025). AI–AI bias: Large language models favor communications generated by large language models. Proceedings of the National Academy of Sciences, 122(31), e2415697122. DOI: https://doi.org/10.1073/pnas.2415697122

[9] Guha, S., Khan, F. A., Stoyanovich, J., & Schelter, S. (2024). Automated data cleaning can hurt fairness in machine learning-based decision making. IEEE Transactions on Knowledge and Data Engineering, 36(12), 7368–7379. DOI: https://doi.org/10.1109/TKDE.2024.3365524

[10] Guillen-Aguinaga, M., Aguinaga-Ontoso, E., Guillen-Aguinaga, L., Guillen-Grima, F., & Aguinaga-Ontoso, I. (2025). Data quality in the age of AI: A review of governance, ethics, and the FAIR principles. Data, 10(12), 201. DOI: https://doi.org/10.3390/data10120201

[11] Rumiantsau, M., & Fokeev, I. (2026). Semantic Layers for Reliable LLM-Powered Data Analytics: A Paired Benchmark of Accuracy and Hallucination Across Three Frontier Models. arXiv preprint arXiv:2604.25149.

Downloads

Published

16-11-2026

Issue

Section

Articles