我們嘗試將每位病人的病歷做分群，而每一群可以呈現出該群所代表的意義，這樣的結果將使醫生在看病時能更快的了解病人曾經生過哪種類型的病，且因為先將病歷做分群，能更準確的找到類似的病歷，而醫生將能從這些病歷分群的結果來參考治療方法、用藥等等。
由於一份病歷會記載多個疾病，可能會被分到多個群中，所以我們提出了先將病歷切割成多個子病歷，而每個子病歷都代表一種病，最後再將子病歷做分群，如此可以達到將一份病歷分在多群的目的，此分群法會比直接將病歷做可重疊分群演算法(overlap clustering algorithm)還來的準確，而且我們可以了解每一群中的病歷是因為有共同或相似的醫學概念(medical concept)被分在同一群。

We try to develop a method to cluster medical records and further know the meaning of each cluster. We hope that the output of the method we developed in this paper can provide a doctor with a good way to quickly understand the disease history of a patient. Furthermore, a doctor can use the output as a reference to find a best treatment to the patient.
Since a medical record may record a variety of diseases, it may belong to multiple clusters. Therefore, we propose a clustering algorithm which will split the medical records into the sub-medical records first. Each sub-medical records can be considered as a medical record that records a single disease. Then we will group related sub-medical records into a cluster. Finally, a medical record is regarded as belonging to the clusters where its sub-medical records belongs to.

第一章 簡介 1
第二章 病歷概念擷取 2
2.1 UMLS(Unified Medical Language System) 2
2.2 MetaMap[8] 2
2.3 醫學縮寫表建立 3
2.4 語義類別(Semantic Type)挑選 3
2.5 一詞對應多個醫學概念 4
2.6 醫學概念表(Concept List) 4
第三章 病歷內概念分群 5
3.1 Jarvis Patrick分群演算法 5
3.2 修改Jarvis Patrick分群演算法 5
3.3 醫學概念相似度計算 6
3.3.1 找上義字 6
3.3.2 關聯性 8
3.3.3 相似度計算 10
3.4 醫學概念分群 12
第四章 病歷分群 13
4.1 病歷切割 13
4.2 病歷相似度計算 13
4.3 病歷分群 15
第五章 實驗結果 17
5.1 人工擷取醫學概念 17
5.2 分群結果 20
5.3 分群結果分析後討論 24
第六章 結論 26
參考文獻 27

[1] “AAMC-specific abbreviations approved for use in the medical record”, Anne Arundel Medical Center, 2017. [Online]. Available: http://studylib.net/doc/8111679/aamc-specific-abbreviations-approved-for-use-in-the-medic.... [Accessed: 09-Aug-2017].
[2] L. Plaza, A. Diaz, and P. Gervas, “A semantic graph-based approach to biomedical summarisation”, Artificial Intelligence in Medicine, vol. 53, no. 1, pp. 1-14, Sep, 2011.
[3] N. Limsopatham, C. Macdonald, and I. Ounis, “Inferring conceptual relationships to improve medical records search”, OAIR, pp. 1-8, May, 2013.
[4] R. Jarvis and E. Patrick, “Clustering Using a Similarity Measure Based on Shared Near Neighbors”, IEEE Transactions on Computers, vol. C-22, no. 11, pp. 1025-1034, Nov, 1973.
[5] “Unified Medical Language System (UMLS)”, National Library of Medicine, 2017. [Online]. Available: https://www.nlm.nih.gov/research/umls/. [Accessed: 18-July-2017].
[6] “Metathesaurus – Rich Release Format(RRF)”, National Library of Medicine, 2017. [Online]. Available: https://www.ncbi.nlm.nih.gov/books/NBK9685/. [Accessed: 18-July-2017].
[7] “UMLS – Metathesaurus Release Abbreviations”, National Library of Medicine, 2017. [Online]. Available: https://www.nlm.nih.gov/research/umls/knowledge_sources/metathesaurus/release/abbreviations.html. [Accessed: 18-July-2017].
[8] “MetaMap – A Tool For Recognizing UNLS Concepts in Text”, National Library of Medicine, 2017. [Online]. Available: https://metamap.nlm.nih.gov/. [Accessed: 18-July-2017].
[9] J. Bezdek, R. Ehrlich and W. Full “FCM: Fuzzy c-means clustering algorithm”, Computers & Geosciences, vol. 10, no. 2-3, pp. 191-203, 1984.
[10] R. Suganya, R. Shanthi, “Fuzzy C- Means Algorithm- A Review”, Int. J. Sci. Res. Publ., vol. 2, no. 11, pp.1-3, Nov. 2012.
[11] N. Grover, “A study of various Fuzzy Clustering Algorithms”, International journal of Engineering Research, vol. 3, no. 3, pp. 177-181, 2014.
[12] N. Pal, K. Pal, J. Keller and J. Bezdek, “A possibilistic fuzzy c-means clustering algorithm”, IEEE Transactions on Fuzzy Systems, vol. 13, no. 4, pp. 517-530, Aug. 2005.
[13] R. Krishnapuram and J. Keller, “A Possibilistic Approach to Clustering”, IEEE Transactions on Fuzzy System, vol. 1, no. 2, pp.98-110, May. 1993.
[14] J. Liu, “Fuzzy modularity and fuzzy community structure in networks”, The European Physical Journal B, vol. 77, no. 4, pp. 547-557, 2010.