Comprehensive Evaluation Method of Urban Water Resources Utilization Based on Dynamic Reduct

As a pre-requisite for optimal urban water resource management, the challenge of devising a comprehensive evaluation method of urban water resource utilization has for years caught the attention of researchers and stakeholders. Yet, even when dealing with the complicated water resource management system comprising of large amounts of vague and incomplete data, an accurate assessment or useful indicators obtained from such data can be of great assistance to municipal managers and policy makers. This article presents a DRAEW algorithm based on (F-lambda)-dynamic reduct method for generating evaluation rules from historical local water environmental data. The proposed method introduces a conditional probability classifier for equivalence class and (F-lambda)-dynamic reduct method for feature extraction and decision rules generation to classify ambiguous undecided objects. The approach is tested using data from Beijing whose water scarcity problem not only represents a microcosm of China but generally illustrates the value of the new approach. The result based on DRAEW algorithm brings the provisional conclusion that Beijing's water status is likely to decline, a result which could greatly assist policy makers and water managers make better decisions.