本論文目的為發展出一套適用於「含水效應物體」的介電常數量測兼相似度辨識預測系統，結合過去介電常數量測方法，又增添「統計相似度分析」和「機器學習的預測」兩種技術去發展，此系統將來可以應用在「所有含水效應的預測兼辨識」。
此文主要是利用開放式同軸探針法量測技術搭配相似度辨識和機器學習的預測來構成。大綱為利用網路分析儀（Vector Network Analyzer, VNA）經由量測探針發射出電磁波訊號，而電磁波訊號由VNA傳輸出去後會與待測物體發生碰撞，碰撞後即會產生反射係數 及透射係數 兩種係數。由VNA測得反射係數 後，再將數據套用在相似度方法進行相似度比對。最後因為經費關係，故機器學習的訓練數據是取材自參考文獻[1]，搭配機器學習模型預測出介電常數（Dielectric Constant）的實部及虛部，如此將可利用此預測系統省下昂貴的量測設備及量測所需要的時間。

The purpose of this paper is to develop a set of similar and predictive systems for the identification and predictive of the dielectric constant applicable to "objects of Water-content Effect". Combining the methods of measurement of dielectric constant in the past, the addition of "statistical similarity analysis" and "machine learning forecasting technology ", which can be used in the identification and prediction of all objects of water content components" in the future.
This paper mainly used open-coaxial probe method to collocation the similarity degree identification and machine learning predict model to composition. The outline is through measurement probe of Vector network analyzer (VNA) to emit electromagnetic signals, and the collision with the object to be measured after electromagnetic signals from the VNA that will be transmitted out. After the collision will produce two coefficients, reflection coefficient(S_11)and transmission coefficient(S_21). After the reflection, coefficient(S_11) is measured by VNA, and the data is applied to the similarity method for similarity comparison. Finally, because of the lack of funds relationship, the training data of the machine learning is derived from the reference [1], with the machine learning model to predict the dielectric constant of the real and imaginary parts, so the future we will be able to use this prediction system to save Expensive measurement equipment and the time required for measurement.

參考文獻
[1] M. T. Hallikainen, F. T. Ulaby, M. C. Dobson, M. A. El-Rayes, and L.-K. Wu, "Microwave dielectric behavior of wet soil-part 1: Empirical models and experimental observations," IEEE Transactions on Geoscience and Remote Sensing, pp. 25-34, 1985.
[2] A. Darvishi, "Translation invariant approach for measuring similarity of signals," Journal of Advances in Computer Research, vol. 1, pp. 19-27, 2010.
[3] P. Chen, H. Ma, S. Gao, and Y. Huang, "SSL: signal similarity-based localization for ocean sensor networks," Sensors, vol. 15, pp. 29702-29720, 2015.
[4] C. Cassisi, P. Montalto, M. Aliotta, A. Cannata, and A. Pulvirenti, "Similarity measures and dimensionality reduction techniques for time series data mining," in Advances in data mining knowledge discovery and applications, ed: InTech, 2012.
[5] C. E. Akbaş, O. Günay, K. Taşdemir, and A. E. Çetin, "Energy efficient cosine similarity measures according to a convex cost function," Signal, Image and Video Processing, vol. 11, pp. 349-356, 2017.
[6] H. L. Kennedy, A new statistical measure of signal similarity: IEEE, 2007.
[7] 戴裕國, "微波頻帶之介質常數量測 " 碩士, 電機工程系, 國立臺灣大學, 1990.
[8] 李政達, "微波介電材料之量測分析與其於天線上之應用 " 碩士, 電信工程學系, 大葉大學, 2006.
[9] 黃微萍, "微波於土壤遙測之研究," 碩士, 太空所, 中央大學, 1994.
[10] A. Henze, N. Tempone, G. Monasterios, and H. Silva, "Incomplete 2-port vector network analyzer calibration methods," in Biennial Congress of Argentina (ARGENCON), 2014 IEEE, 2014, pp. 810-815.
[11] 黄斌, 魏强, 杨克勇, and 矣峻岭, "矢量网络分析仪十二项误差校正模型理论推导," 电子工业专用设备, vol. 9, p. 005, 2009.
[12] 林宗勳, "Support Vector Machines 簡介," ed: 台灣大學通訊與多媒體實驗室, 2006.
[13] 黃子桓, ""Support Vector Regression"."
[14] 資管系, "隨機森林運用於白血病基因分類," in KC 2013 第九屆知識社群研討會論文集, 2013, pp. 112-122.
[15] B. O’Connor, "Cosine similarity, Pearson correlation, and OLS coefficients," AI and Social Science Blog. Web. Accessed May, vol. 12, p. 2015, 2012.
[16] 易顺明, "基于 Python 的推荐系统相似性分析和协同过滤," 沙洲职业工学院学报, pp. 3-7, 2015.
[17] A. Bellet, "Tutorial on Metric Learning," 2013.
[18] R. De Maesschalck, D. Jouan-Rimbaud, and D. L. Massart, "The mahalanobis distance," Chemometrics and intelligent laboratory systems, vol. 50, pp. 1-18, 2000.
[19] A. Bajoulvand, R. Z. Marandi, M. R. Daliri, and S. H. Sabzpoushan, "Analysis of folk music preference of people from different ethnic groups using kernel-based methods on EEG signals," Applied Mathematics and Computation, vol. 307, pp. 62-70, 2017.
[20] D. Wang, D. S. Yeung, and E. C. Tsang, "Weighted mahalanobis distance kernels for support vector machines," IEEE Transactions on Neural Networks, vol. 18, pp. 1453-1462, 2007.
[21] L. J. Bain and M. Engelhardt, Introduction to probability and mathematical statistics: Brooks/Cole, 1987.
[22] M. I. Ribeiro, "Gaussian probability density functions: Properties and error characterization," Institute for Systems and Robotics, Lisboa, Portugal, 2004.
[23] J. Behari, Microwave dielectric behaviour of wet soils vol. 8: Springer Science & Business Media, 2006.
[24] 黃進芳, "微波工程," 提昇大學基礎教育計畫寬頻通訊系統基礎教育計畫, 國立台灣科技大學2003.
[25] C.-J. Huang, "介電係數量測技術之研究," National Central University, 2002.
[26] 叶荣芳 and 徐佳, "SOLT 校准方法及其在射频测量中的应用," 電子器件, vol. 29, pp. 179-182, 2006.
[27] 李岳璁 and 林漢年, "量測液體介電參數之終端開路同軸探棒的分析與設計," 撰者, 2004.
[28] 葉昭緯 and 張振雄, "利用一維光子晶體對低吸收材料的介電性質, 厚度之特性檢測與探討," 2006.
[29] J. Baker-Jarvis, Transmission/reflection and short-circuit line permittivity measurements: National Institute of Standards and Technology Colorado, 1990.
[30] J. Shenhui, D. Ding, and J. Quanxing, "Measurement of electromagnetic properties of materials using transmission/reflection method in coaxial line," in Environmental Electromagnetics, 2003. CEEM 2003. Proceedings. Asia-Pacific Conference on, 2003, pp. 590-595.
[31] 张庆辉, 王继红, and 金广锋, "基于自由空间法的粮堆介电特性测试研究," 河南工業大學學報 (自然科學版), vol. 31, pp. 66-68+ 75, 2010.
[32] N. N. Rao, Elements of engineering electromagnetics: Prentice Hall, 1977.
[33] 謝劍斌、興軍亮、張立甯、方宇強、李沛秦、劉通、閆瑋、王勇沈傑、張政、譚筠、胡俊, 改變未來20年最重要的20個視覺機器學習理論深讀, 2016.
[34] T. M. Mitchell, "Machine learning. 1997," Burr Ridge, IL: McGraw Hill, vol. 45, pp. 870-877, 1997.
[35] P. Sastry, "An introduction to support vector machines," Computing and information sciences: Recent trends, pp. 53-85, 2003.
[36] D. Basak, S. Pal, and D. C. Patranabis, "Support vector regression," Neural Information Processing-Letters and Reviews, vol. 11, pp. 203-224, 2007.
[37] N. Cristianini and J. Shawe-Taylor, An introduction to support vector machines and other kernel-based learning methods: Cambridge university press, 2000.
[38] L. E. Peterson, "K-nearest neighbor," Scholarpedia, vol. 4, p. 1883, 2009.
[39] D. SaedSayad, "K Nearest Neighbors," University of Toronto2010.
[40] X. Zhu, P. Gong, Z. Zhao, and C. Zhang, "Learning similarity metric with SVM," in Neural networks (ijcnn), the 2012 international joint conference on, 2012, pp. 1-8.
[41] 謝璦如, "分類與回歸樹演算法," 中國醫藥大學 生物統計中心.
[42] B. Chen, "Decision Tree Learning."
[43] 陳士杰, "決策樹學習," 聯合大學資訊管理學系2005.
[44] CH.Tseng. (2017). 隨機森林Random Forest. Available: https://chtseng.wordpress.com/2017/02/24/%E9%9A%A8%E6%A9%9F%E6%A3%AE%E6%9E%97random-forest/