醫療對於現今社會來說是重要的一部分，更是與生活有著密不可分的關係，人們總是希望能盡量減少到醫院的次數，也總是盼著候診的等待時間更是能夠縮短。因為各種因素，等待時間也會有所不同，尤其像是在大型醫院，更是讓人無法捉摸，如何能夠有效的預測時間，也逐漸變成大家在意的課題。本研究利用診所過去看診的資料，藉由看診時間與各項因素，透過機器學習的方式，建立預測的模型並將預測出來的結果可作為診所管理者改善候診時間之參考。藉由大量的病人預約門診的看診資料，包括時段、科別語科室資訊，以及實際就診時間。將這些數據分為訓練集和測試集，並使用線性回歸 (Linear Regression) 與極限梯度提升 (eXtreme Gradient Boosting) 兩種方式進行模型訓練和預測。本研究目標為減少病人的等待時間，提高就診效率。準確預測候診時間可以幫助醫療機構合理安排醫生和適當的資源，除了可改善就診體驗、提高醫療效率並提供更好的醫療服務。此外，對於未來展望也提供參考，可進一步改進預測模型的準確性和可靠性。 Medical care is an essential and inseparable part of modern society. People always hope to minimize their visits to hospitals and eagerly anticipate shorter waiting times for appointments. However, various factors contribute to different waiting times, particularly in large hospitals, making it difficult to predict effectively. The ability to accurately forecast waiting times has become a significant concern for many individuals.In this study, we utilized historical clinic appointment data, considering factors such as appointment time and various variables related to clinic operations. Through machine learning techniques, we aimed to establish a predictive model that can serve as a reference for clinic managers to improve waiting times. By utilizing a large dataset of patient appointments, including time slots, department information, and actual visit durations, we divided the data into training and testing sets. We employed two methods, namely linear regression and extreme gradient boosting, for model training and prediction.The objective of this research is to reduce patient waiting times and enhance the efficiency of medical care. Accurate prediction of waiting times can assist healthcare institutions in allocating doctors and appropriate resources efficiently. This not only improves the patient experience and overall medical efficiency but also facilitates the provision of better healthcare services. Furthermore, this study provides insights for future improvements, aiming to enhance the accuracy and reliability of predictive models.
