跳到主要內容 :::

年度	113	專案性質	實驗性質
專案類別	研究專案	研究主題	自訂
申請機構	東海大學	申請系所	環工系
專案主持人	陳鶴文	職等/職稱	特聘教授
專案中文名稱	建立以大數據分析與生成式人工智慧為基礎的智慧化整治費管理技術
中文關鍵字	大數據分析,生成式人工智慧,智慧化整治費管理技術
專案英文名稱	Establish intelligent governance fee management technology based on big data analysis and generative artificial intelligence.
英文關鍵字	Big data analysis, generative artificial intelligence, intelligent governance fee management technology.
執行金額
執行期間	2024/12/1 至 2025/11/30
計畫中文摘要	近年來，土壤及地下水整治費管理因產業屬性不同、污染屬性不同、產品製程不同，導致業者在申報時，因不熟環境法規或政策時，不知如何申報整治費，進而需要花費業者與承辦的大量時間，進行協助申報；除此之外，當整治費的資料量大量增加時，管理者要從大量資料中萃取關鍵因素及了解整治費資料所隱含的知識，需要耗費相當多時間與人力成本。為精進且減輕政府部門在整治費管理的時間，近年來，大數據分析與知識管理技術被許多業者所使用，為此，本專案將針對整治費建立大數據分析與人工智慧為基礎的智慧化整治費管理技術，該大數據分析技術主要整合整治費申報系統及營業事業登記資訊、建立大數據分析模組、協助承辦人員快速理解整治費申報關鍵因素並分析產業應繳納相關費用合理性。文本分析的目的是利用「人工智慧技術」針對數據分析的結果，透過深度學習模型的方式，產出申報資料之相關內容。本專案希冀此整合技術可提供政府部門在整治費管理上作為擬定策略工具之一。
計畫英文摘要	In recent years, the management of soil and groundwater remediation fees has faced challenges due to differences in industry attributes, pollution characteristics, and production processes. These differences often result in difficulties for enterprises when reporting remediation fees, particularly if they lack familiarity with environmental regulations or policies. This situation frequently requires substantial time and effort from both enterprises and handling personnel to facilitate the reporting process. Additionally, as the volume of remediation fee data increases, extracting key factors and uncovering the implicit knowledge within the data becomes a time-consuming and labor-intensive task for administrators.To enhance efficiency and reduce the time required by government departments in managing remediation fees, big data analytics and knowledge management technologies have been increasingly adopted by various industries in recent years. Therefore, this project aims to develop an intelligent remediation fee management system based on big data analytics and artificial intelligence. The proposed big data analytics technology will integrate the remediation fee reporting system with enterprise registration information, establish big data analysis modules, and assist personnel in quickly understanding the key factors for remediation fee reporting while evaluating the rationality of associated industry-related fees. The text analysis component leverages "artificial intelligence technologies" to process the results of data analysis through deep learning models, generating relevant content for reporting data. This project aspires to provide an integrated technology solution that can serve as a strategic tool for government departments in the management of remediation fees.