Research combining deep learning methodologies with Analytic Hierarchy Process establishes comprehensive frameworks for data security risk assessment, achieving systematic evaluation and preventive maintenance, while enterprise-scale implementations demonstrate practical applications in financial database management and predictive risk forecasting systems.
-- The rapid expansion of digital infrastructure has heightened data security risks across sectors. Traditional assessment methods, often reliant on fragmented evaluations and reactive maintenance, struggle to meet the complexity of modern threat environments. Recent research published in Procedia Computer Science introduces an integrated framework that combines deep learning technologies with the Analytic Hierarchy Process to support automated risk identification and multidimensional security evaluation.
The model is built around two core components that work together: deep learning is used to automatically extract potential security risks from large volumes of operational data, while the Analytic Hierarchy Process provides structured, expert-weighted evaluation across multiple dimensions. The assessment framework incorporates eight key indicators, including data classification and identification, user authentication and authorization, encryption, backup and recovery, log monitoring, access control, network security management, and lifecycle governance. Consistency testing of the AHP judgment matrix confirms the reliability of the weighting structure, and the fuzzy comprehensive evaluation method is applied to address uncertainty within the scoring process and convert qualitative assessments into quantitative results.
Experiments show that the proposed model performs stably in practical scenarios and improves the accuracy of risk assessment. When applied to a digital financial platform, the framework not only confirmed an overall low level of data security risk but also pinpointed specific weaknesses in encryption, authentication, and access control that conventional assessments tended to overlook. By comparing three targeted remediation strategies using the TOPSIS multi-criteria method, the study identified the option that best balances cost, implementation speed, and expected effectiveness. This process illustrates how the model can guide organizations from risk diagnosis to actionable improvement planning with greater clarity and precision.
The study is authored by Lingyun Lai, who holds a Master’s degree in Enterprise Risk Management from Columbia University and a Bachelor’s degree in Finance from Wenzhou-Kean University. Her academic training and professional experience in financial analysis, vendor risk assessment, and real-time monitoring of over 20 million dollars in project exposure provide a practical foundation for her work in data-driven risk evaluation.
Beyond her academic foundation, Lai has led multiple enterprise-level risk and data systems projects that demonstrate her ability to translate analytical models into measurable outcomes. Her work at BCG Glass Industry Inc. includes designing an integrated financial and risk database covering more than 30 active projects, developing early-warning mechanisms that predict delays and financial anomalies, and building real-time data-quality tools that reduce reporting errors by over 40 percent. She has also authored peer-reviewed research in financial modeling, AI valuation, and data-driven credit risk assessment, with a growing citation record reflecting her interdisciplinary contributions. These initiatives highlight her capacity to merge quantitative risk theory with practical system design across construction, advanced manufacturing, and financial services.
Taken together, Lai’s research and practical work show how data-driven risk assessment can be applied in real organizational settings. The model demonstrated in the digital financial platform case provides a structured way to identify risks and support corrective actions, while her experience building financial risk databases, early warning tools, and real-time data quality systems reflects the broader impact of these methods in enterprise environments.
Contact Info:
Name: Lingyun Lai
Email: Send Email
Organization: Lingyun Lai
Website: https://scholar.google.com/citations?hl=en&view_op=list_works&authuser=3&gmla=AKzYXQ2YBoCZgT-hAv3A2o3zLfHPbcsAuSGM0IJF2KrbbsqaDFfhh37CBLVPPlVp-rFTusv-DUDjpOcs7775zw&user=CD9p49gAAAAJ
Release ID: 89176890
If you come across any problems, discrepancies, or concerns related to the content contained within this press release that necessitate action or if a press release requires takedown, we strongly encourage you to reach out without delay by contacting error@releasecontact.com (it is important to note that this email is the authorized channel for such matters, sending multiple emails to multiple addresses does not necessarily help expedite your request). Our committed team will be readily accessible round-the-clock to address your concerns within 8 hours and take appropriate actions to rectify identified issues or support with press release removals. Ensuring accurate and reliable information remains our unwavering commitment.
