本研究主要考慮在盡量保留可讀性的情況下，找出將 Imperative Programming 程式碼對應的 Functional Programming 的程式碼並證明。 結果如下： 一、if statement 等價於由 ifte 函數所構成的程式碼，其中函數ifte定義在本文內 二、某些 for-loop statement 等價於由 foldl 函數所構成的程式碼 三、某些 for-loop statement 等價於由 map 函數所構成的程式碼

Max Gold's project, titled “Solving Mathematical and Chemical Equations using Python”, is a website comprising of 4 main programmes: one to find the smallest possible combination of two chemical compounds or elements; a self-made parsing function to convert a chemical equation into a matrix, then using Gaussian-Jordan elimination to find coefficients for an equation; a programme to parse a mathematical expression and use that parsed expression in algebraic division of an algebraic dividend of nth degree polynomial by a divisor of 1st degree polynomial; finally, a programme to solve binomial equations for the power s∈Q. This website was originally made so that Max Gold could improve his programming skills for GCSE computer science but expanded to incorporate his passion for chemistry and maths and thus allow others to use these programmes to help them with their problems as well. A problem with many conventional calculator websites is their lack of specificity – they tend to be able to compute some functions but not all. These programmes are tailored to GCSE and A level maths and chemistry, meaning this website provides an outlet to compute specific topics of problems.

我們設計一AIoTs物聯網相變實驗系統，觀察與分析多體粒子的彈跳運動行為，並以二分法來觀察系統在不同驅動電壓下的粒子數失衡參數α與有序參量β；在改變頻率實驗中，觀察到低驅動電壓下的粒子集體行為(collecting behavior)、臨近相變臨界點之粒子漸近行為(Approaching behavior)、與遠離相變臨界點之粒子漲落行為(Fluctuation behavior)皆與頻率有關，且臨界電壓隨頻率增加，呈現指數遞減，約在6 Hz後即趨於常數3.6 eV，而高驅動電壓下之粒子漲落行為仍依然存在，而量子電腦計算結果說明，利用26-Qubits成功描述20個自旋粒子在控制旋轉閘(Controlled Rotate Y-gate)運算下的相變行為，並能完美擬合相變區間之漸進行為(Approaching behavior)。藉由本次科展所提出之實驗模型，可以在未來為凝體物理系統與相變實驗提供更豐富的研究課題與更廣闊的研究前景。

本研究旨在探討「水滴的帶電程度」，利用攝影機拍攝水滴受靜電棒作用後的現象，搭配Tracker電腦軟體，觀察「各種水滴」受「不同靜電力」的影響所產生「偏移」、「裂解」、「單位水滴質量的電量負載(q/m)」...等結果有何差異。研究結果發現：(一)水滴的確會帶電，並可繞轉或吸附靜電棒；(二)水滴受靜電作用後會在針尖發生裂解，裂解後的小水滴極限質量為0.00085(克)；(三)從水滴的軌跡可得知(加)速度，透過x-t、v-t與a-t圖，可推知水滴裂解瞬間的(加)速度、重力與靜電力間的關係，進而估算「q/m」，分析水滴的帶電程度；(四)改變水滴的類型並重複實驗可知：鹽水的偏移量、帶電量、荷質比最大，而酒精最小；因此，帶電離子與氫鍵可能影響水滴的帶電程度。

近年來原住民語使用逐漸減少，面臨傳承的危機。本研究製作太魯閣族語音評分系統，希望幫助人們學習太魯閣族語。利用原委會網站的音檔和田野調查收集錄製的族語語音與族語老師的評分為本研究的研究語料，以Kaldi為語音辨識框架來製作語音辨識模型。對模型輸出的音素與正確單詞進行字串相似度計算，將計算數據與模型產出的發音品質分數作為選擇模型的依據，透過機器學習方法選擇最接近族語老師評分的模型。以此模型計算研究語料，再以機器學習製作一個七成正確率的評分分類器，最後實作出評分系統網站。此網站除了可以學習族語之外，也可以幫助未來繼續收集更多語料，提高評分系統的正確率。期望幫助人們學習太魯閣族語，讓原住民族語原音得以重現。

本研究旨在發展一套可限制特徵之人臉生成器，能依據目擊者對嫌犯的特徵描述，使用真實人臉照片進行特徵生成，以生成嫌犯臉孔。本研究以生成對抗網路為中心，配合選擇性傳輸單元、編碼器與解碼器結構，發展出五官、毛髮、飾品、整體形象四大類別，共有40個特徵生成編輯種類。搭配友善的圖形使用者介面，讓使用者可將已相關度篩選後之真實人臉資料庫或是另行輸入之照片，因應不同的嫌犯臉孔描繪需求情境，對該圖片臉孔進行指定特徵之生成編輯。此技術除了可輔助刑案目擊者回憶嫌犯臉孔特徵，亦可應用於眼鏡業、醫學美容、美髮業等領域。

在資訊爆炸的時代，效率閱讀、整理資料的能力越趨重要。身為高中生，學習時的閱讀量龐大，還須另外自己挑選重點句，重新整理筆記。因此我想如果可以讓電腦自動摘取文章的重點，就能幫助學生效率學習。 大多數現存的自動摘要研究適用於英文文本，本研究利用演算法抓取中文文章的摘要，使學生可以真正實用該演算法於日常學習當中。除此之外，此研究比較了不同方法摘要的準確率以及優缺點。

全球貧血人口普遍，然許多人並不了解自身是否罹患貧血；長期患有貧血的病人，亦需定期抽血檢驗追蹤是否有貧血惡化達到需接受輸血的程度。研究顯示，結膜之顏色與貧血有絕對關係，結膜越白則貧血越嚴重，醫師也常使用結膜顏色推測是否有貧血情形。若能設計手機軟體自動分割結膜影像並分析其顏色，將有機會推測受試者是否罹患貧血。本研究收集22位無貧血者及8位貧血病人，並獲得其近期血紅素數值。以手機取得受試者之眼睛影像後，成功設計程式以深度學習完成結膜自動影像分割，對於分割影像以面積大小進行後期處理後，依其取得下眼瞼結膜之三原色平均，再利用kNN與SVM演算法判斷預測出該受試者是否具有貧血之症狀。本研究主要分為兩階段，其一為進行下眼瞼結膜分割模型訓練；其二為製作有無貧血之判斷模型。整合上述眼瞼分割模型(IoU=89.8%±0.02%)與貧血判斷模型(SVM以polynomial核函數測出 準確值93.3%±24.3%)後，可得貧血診斷準確率為80%。此結果代表AI技術有機會透過結膜影像，判斷被拍攝者是否有貧血情形，未來若能增加研究人數，將可設計網頁版或手機APP加以推測血紅素值，供大眾居家篩檢。

We study the creation of physical adversarial examples, which are robust to real-world transformations, using a limited number of queries to the target black-box neural networks. We observe that robust models tend to be especially susceptible to foreground manipulations, which motivates our novel Foreground attack. We demonstrate that gradient priors are a useful signal for black-box attacks and therefore introduce an improved version of the popular SimBA. We also propose an algorithm for transferable attacks that selects the most similar surrogates to the target model. Our black-box attacks outperform state-of-the-art approaches they are based on and support our belief that the concept of model similarity could be leveraged to build strong attacks in a limited-information setting.

做研究時，經常需要閱讀多篇長篇論文，如何快速理解並歸納重點是影響效率收益的關鍵任務。針對此問題本研究運用深度學習配合其他程式算法輔助。將整個問題拆解成三個主要部分：模型處理、結構剖析、心智圖輸出，模型處理的部分我們訓練了三種不同的Seq2seq模型實現文本摘要，分別使用LSTM、BERT、ALBERT、比較並選擇綜合效益最好的模型；而負責拆解論文的結構剖析系統，則利用論文目錄分解整份論文，最後，心智圖輸出系統則整合剖析系統與模型結果，再調用Xmind API去生成架構清晰易讀易理解的論文心智圖。

Recent studies showed that the probability of Taiwanese females developing breast cancer has risen dramatically over the past 30 years. We are now facing younger and more breast cancer patients in Taiwan. What makes the matter even more severe, is the fact that patients that take cancer treating medicine will suffer from its serious side effects, some may even lose the ability to reproduce. We hope to develop a new system that can help doctors and researchers develop new medicine for treating breast cancer, the way medicine cures cancer tumors are by attaching onto the infected cells’ receptors. After collecting MACCS data (converted from SMILES), the dataset will be used for training the machine learning program. Due to the problem of insufficient training data, we used an ensemble method to generate our machine learning model. Among the three basic ensemble techniques, Max Voting, Averaging, and Weighted Averaging. we selected the max voting technique to perform the prediction for this research. We created two separate datasets, positive and negative, the two datasets will later be used as training data for the program. We weren’t sure of the ratio of positive and negative in the training data, therefore we compare 40 different ratios and evaluate the results. By comparing the accuracy of the models, we found out that when the ratio between positive data and negative data is 1:3000, the machine learning program will have the highest precision. After we created the final model through voting among the 1000 models generated, we evaluate the precision of the model through the following methods, AUC, precision, recall. The ultimate goal of this research is to assist doctors and researchers shorten the process of developing and testing new medicines.

Triage is the process by which nurses manage hospital emergency departments by assigning patients varying degrees of urgency. While triage algorithms such as the Emergency Severity Index (ESI) have been standardized worldwide, many of them are highly inconsistent, which could endanger the lives of thousands of patients. One way to improve on nurses’ accuracy is to use machine learning models (ML), which can learn from past data to make predictions. We tested six ML models: random forest, XGBoost, logistic regression, support vector machines, k-nearest neighbors, and multilayer perceptron. These models were tasked with predicting whether a patient would be admitted to the intensive care unit (ICU), another unit in the hospital, or be discharged. After training on data from more than 30,000 patients and testing using 10-fold cross-validation, we found that all six models outperformed ESI. Of the six, the random forest model achieved the highest average accuracy in predicting both ICU admission (81% vs. 69% using ESI; p<0.001) and hospitalization (75% vs. 57%; p<0.001). These models were then added to an Android application, which would accept patient data, predict their triage, and then add them to a priority-ordered waiting list. This approach may offer significant advantages over conventional triage: mainly, it has a higher accuracy than nurses and returns predictions instantaneously. It could also stand-in for triage nurses entirely in disasters, where medical personnel must deal with a large influx of patients in a short amount of time.