「非都市土地開發審議」案例式潛在問題警示系統

A Case Based Problem Alert System for Land Development Approval

黃佩菁(Pei-Jing Huang)

指導教授 ： 林峰田

國立臺灣大學/工學院/建築與城鄉研究所/碩士(2004年)

https://doi.org/10.6342/NTU.2004.01613

摘要

現行相關非都市土地開發審議法令多屬通則性規範，常有因法令定義不清或待解釋而產生之問題。因此，本研究欲運用過去審議之經驗，結合案例式推理之原理，來建構出一套輔助系統。當審議人員面臨新的開發案件時，可透過系統的推論，從過去的案例中找出情況相似的案例，而這些案例便可以提醒審議人員可能會產生的問題是什麼。
本系統在EXCEL環境下執行。使用者依照屬性填入屬性值並設定情境相似性門檻值、權重後，系統便會自動進行比對找出相似的案例，當所推論出的情境相似性小於或等於所設定之門檻值，即表示此案例的考量因子有助於提醒審議人員可能會產生問題的地方為何。
本研究以最近區域計畫委員會審議的33個案件建構案例庫，並據以求得最適之情境相似性門檻值及權重組合，使得有最高的有效率及最低的誤判率。

關鍵字

土地開發審議 ； 案例式推理

並列摘要

Many land development scrutiny ordinances are stated in terms of general principles which may be vague and cause scrutiny problems. Employing the technique of case-based reasoning, this research develops an alert system to assist the project reviewers to find potential problems based on the experiences of previously similar cases.
This system is implemented by EXCEL. After the users give attribute values of the case in question, set a threshold value defining the range of similarity, and weights of every attributes, the system will find similar cases from previous cases automatically. A case associated with a value which is smaller or equal to the given threshold is considered a similar case. Problems found in the similar cases will be prompted to alert the users.
This research takes 33 cases which have completed their scrutiny processes by the Regional Planning Committee to built up a case base. Accordingly, a set of appropriate thresholds and weights with highest probability of useful cases and lowest barren ratios.

並列關鍵字

land development approval ； case-based reasoning

參考文獻

林建元. (1992). 山坡地土地利用管理制度之建立﹝第二期﹞, 台灣大學建築與城鄉研究所.

Bareiss, E. R., and Poter, B. W. (1988). "Protos: An Exemplar-Based Learning Apprentice." International Journal of Man-Machine Studies, 29, 549-561.

Bergmann, R. (1998). "Introduction to Case-Based Reasoning." Artificial Intelligence Communications, 7, 23-39.

Carver, S. J. (1991). "Integrating Multi-Criteria Evaluation with Geographic Information." International Journal of Geographical Information System, 5, 321-339.

Garver, S. J. (1991). "Integrating Multi-Criteria Evaluation with Geographic Information." International Journal of Geographic Information System, 5, 321-339.