陳信希臺灣大學：資訊工程學研究所李遲Lee, ChihChihLee2007-11-262018-07-052007-11-262018-07-052005http://ntur.lib.ntu.edu.tw//handle/246246/53921Named entity recognition is a fundamental task in biomedical text mining. Multiple-class entity annotation is more complicated and challenging than single-class entity annotation. In this thesis, we presented a single word classification approach to dealing with the multiple-class entity annotation problem using Support Vector Machines (SVMs). In other words, each token in a sentence is represented by a feature vector and classified as one of the given classes. Orthographical patterns, morphological patterns, results from existing gene/protein name taggers, context, part of speech (POS) tags, tags (class labels) of surrounding tokens, and other information are important features for named entity recognition. In addition, we employed a unique way of extracting and utilizing context information. Due to the huge number of non-entity instances (class ‘O’), we clustered the instances of this class into 5 subclasses to accelerate the SVM training process. We also applied a simple post-processing technique with the help of a dictionary and a post-processing technique via abbreviation extraction. We presented the performance of our system using 13 different notions of correctness, showing the overall performance of our system is somewhere between 68.16% and 79.91% in terms of f-score, which is comparable to the performance of the top 3 systems in the JNLPBA shared task. Besides various notions of correctness used in evaluation, we defined 5 types of errors and showed how frequently our system made these types of mistakes. The error analysis also revealed the annotation discrepancies among the training and test corpora. Therefore, researchers approaching biomedical named entity recognition with machine learning algorithms should seek to improve their systems as well as be aware of the correctness of the underlying corpus.List of Figures III List of Tables IV 1. Introduction 1 1.1. Named Entity Recognition in Biomedical Literature 1 1.2. Organization of this Thesis 3 2. Training and Test Corpora 5 3. Methods 9 3.1. SVM on Named Entity Recognition 9 3.2. Features 11 3.2.1. Orthographical Patterns 12 3.2.2. Morphological Patterns 12 3.2.3. Information from Gazetteers 15 3.2.4. Context Information 16 3.2.5. Tags of Surrounding Tokens 18 3.2.6. Other Features 21 3.3. Post-Processing Operations 23 3.3.1. Dictionary-Assisted Post-Processing 23 3.3.2. Post-Processing via Abbreviation Extraction 24 4. Results and Discussion 26 4.1. Evaluation Criteria and Metrics 26 4.2. Overview of the Results 28 4.3. Results of Each Subset 33 4.4. Comparison with the Participating Systems in the JNLPBA Shard Task 34 4.5. Error Analysis 35 4.5.1. Correct Boundaries but Class Label 36 4.5.2. Excessive/Missing Leading/Ending Tokens in Proposed Named Entities 37 4.5.3. Concatenation of Named Entities 39 4.5.4. Tagging Parentheses 41 4.5.5. Tagging Coordinating Conjunctions 42 4.6. Discussion 44 5. Conclusion and Future Work 46 References 47446228 bytesapplication/pdfen-US生物資訊自然語言處理生醫具名實體支援向量機bioinformaticsNLPbiomedical named entitysupport vector machinesthesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/53921/1/ntu-94-R92922005-1.pdf