Article: Adaboost algorithm-based cost risk assessment for university laboratory construction Journal: International Journal of Reasoning-based Intelligent Systems (IJRIS) 2026 Vol.18 No.8 pp.10 - 22 Abstract: With the rapid expansion of university laboratories, cost overruns have become a critical issue due to accelerating hardware iterations, rising hidden costs, and significant interdisciplinary disparities. Traditional risk assessment methods, such as multiple linear regression and Monte Carlo simulation, struggle to handle nonlinear interactions and data heterogeneity. To address these challenges, this paper proposes a dynamic weight-adjusted AdaBoost algorithm for cost risk assessment. The approach incorporates a multimodal feature fusion mechanism integrating hardware, software, and implicit cost domains, alongside a domain-knowledge guided weighting strategy. Experimental results on a multi-disciplinary dataset show that the proposed method reduces the mean absolute percentage error by 26.5% and improves the F1-score for high-risk event identification to 0.893, significantly outperforming existing benchmarks. The framework also enables earlier risk warnings and more effective cost control strategies. Inderscience Publishers - linking academia, business and industry through research

Title: Adaboost algorithm-based cost risk assessment for university laboratory construction

Authors: Pengfei Zhao

Addresses: School of Mechatronic Engineering and Automation, Shanghai University, Shanghai, 200444, China

Abstract: With the rapid expansion of university laboratories, cost overruns have become a critical issue due to accelerating hardware iterations, rising hidden costs, and significant interdisciplinary disparities. Traditional risk assessment methods, such as multiple linear regression and Monte Carlo simulation, struggle to handle nonlinear interactions and data heterogeneity. To address these challenges, this paper proposes a dynamic weight-adjusted AdaBoost algorithm for cost risk assessment. The approach incorporates a multimodal feature fusion mechanism integrating hardware, software, and implicit cost domains, alongside a domain-knowledge guided weighting strategy. Experimental results on a multi-disciplinary dataset show that the proposed method reduces the mean absolute percentage error by 26.5% and improves the F1-score for high-risk event identification to 0.893, significantly outperforming existing benchmarks. The framework also enables earlier risk warnings and more effective cost control strategies.

Keywords: AdaBoost; cost risk assessment; university laboratories; multimodal data fusion; dynamic weighting mechanism.

DOI: 10.1504/IJRIS.2026.151724

International Journal of Reasoning-based Intelligent Systems, 2026 Vol.18 No.8, pp.10 - 22

Received: 02 Nov 2025
Accepted: 27 Dec 2025
Published online: 17 Feb 2026 *

Title: Adaboost algorithm-based cost risk assessment for university laboratory construction

Keep up-to-date