自傳

您好，我是陳冠廷，有兩年以上的AI研究經驗以及伺服器相關知識應用，曾負責研究探討數據工程，讓廠商可以方便的撈取資料，繪製圖表進行數據分析。 過往工作除了維護和開發網頁功能外，主要透過統計、Python撰寫ML 及 DL 手法分析電力、病摘資料，建構電力負載預測以及疾病編碼分類預測。

在專題實作中，了解許多相關業界知識並從中獲取實作經驗，藉由大學所學基礎與物聯網技術、工業4.0結合，透過控管水質的各項數據，如PH、ORP、DO、Temp等多個不同感測器，營造最合適魚苗的環境，達成智慧養殖自動化目的，過程中雖然也曾遭遇困難，如封包傳輸格式、封包堵塞造成死機，並藉由RESTful API以及多執行緒方式解決相關問題，但也讓我體會到共同討論與分工合作的重要性，不僅有助於發現問題，更利於團隊間的程式建置與統整。

 研究所期間參與多家廠商的產學合作計畫，如臺大醫院 、再生能源應用，碩一階段的研究主軸為能源負載預測，經由論文研討與同儕、教授討論能源相關議題走向，找出各項潛在影響參數，有效提升研究績效，也藉此機會訓練自身的口語表達能力，並加深對數據分析的專業知識，奠定了機器學習相關基礎知識；碩二階段研究主軸為自然語言(NLP)，參與醫院的產學合作，發表基於醫學擴充詞進行國際疾病編碼之多標籤分類論文及參與研討會，善用不同演算法的特性在詞向量方面，如Word2Vec、FastText，結合卷積神經網路模型進行預測，能有效提取文章中特徵，並導入Attention機制對疾病預測準確率有極大幫助；此外，課餘期間也曾負責網站建置與維護(PHP)等相關工作，並擁有一年相關實作經驗。  

In this project, I learned lots of industry knowledge, practical experience from it. By combining the foundation of university study with IoT technology and Industry 4.0, The main goal is to create the most suitable environment for fish fry by controlling various data of water quality monitor, such as PH, ORP, DO, Temp, etc., . For achieving the purpose of automated aquaculture systems. Although I encountered difficulties in this process such as packet transmission format, packet blockage, and crashes, The way to solve these problems is through RESTful API and multiple executions, and I also recognized the importance of common discussion and division of labor which is not only to help identify problems but also facilitates the establishment of the program and the integration between teams. 

During the research period, we were involved in several industry-academia collaborative projects with manufacturers, such as hospitals and renewable energy applications. The research at the master's level that we focused on energy load forecasting, and through these discussions with peers and professors, we discussed the direction of energy-related issues, identified potential impact parameters, and effectively improved research performance. In addition, my research interests are disease coding classification prediction, using NLP processing to obtain disease word vectors, combined with deep learning models for prediction, effectively improving the accuracy of disease prediction, and also responsible for website development (PHP) and other related work after school, with one year of relevant practical experience.