MARIE: A Context-Aware Term Mapping with String Matching and Embedding Vectors
With growing interest in machine learning, text standardization is becoming an increasingly important aspect of data pre-processing within biomedical communities. As performances of machine learning algorithms are affected by both the amount and the quality of their training data, effective data sta...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-11-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/10/21/7831 |
| _version_ | 1827702862645297152 |
|---|---|
| author | Han Kyul Kim Sae Won Choi Ye Seul Bae Jiin Choi Hyein Kwon Christine P. Lee Hae-Young Lee Taehoon Ko |
| author_facet | Han Kyul Kim Sae Won Choi Ye Seul Bae Jiin Choi Hyein Kwon Christine P. Lee Hae-Young Lee Taehoon Ko |
| author_sort | Han Kyul Kim |
| collection | DOAJ |
| description | With growing interest in machine learning, text standardization is becoming an increasingly important aspect of data pre-processing within biomedical communities. As performances of machine learning algorithms are affected by both the amount and the quality of their training data, effective data standardization is needed to guarantee consistent data integrity. Furthermore, biomedical organizations, depending on their geographical locations or affiliations, rely on different sets of text standardization in practice. To facilitate easier machine learning-related collaborations between these organizations, an effective yet practical text data standardization method is needed. In this paper, we introduce MARIE (a context-aware term mapping method with string matching and embedding vectors), an unsupervised learning-based tool, to find standardized clinical terminologies for queries, such as a hospital’s own codes. By incorporating both string matching methods and term embedding vectors generated by BioBERT (bidirectional encoder representations from transformers for biomedical text mining), it utilizes both structural and contextual information to calculate similarity measures between source and target terms. Compared to previous term mapping methods, MARIE shows improved mapping accuracy. Furthermore, it can be easily expanded to incorporate any string matching or term embedding methods. Without requiring any additional model training, it is not only effective, but also a practical term mapping method for text data standardization and pre-processing. |
| first_indexed | 2024-03-10T15:05:32Z |
| format | Article |
| id | doaj.art-d8af4db735f040e8bb336d19e486921b |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-10T15:05:32Z |
| publishDate | 2020-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-d8af4db735f040e8bb336d19e486921b2023-11-20T19:49:23ZengMDPI AGApplied Sciences2076-34172020-11-011021783110.3390/app10217831MARIE: A Context-Aware Term Mapping with String Matching and Embedding VectorsHan Kyul Kim0Sae Won Choi1Ye Seul Bae2Jiin Choi3Hyein Kwon4Christine P. Lee5Hae-Young Lee6Taehoon Ko7Office of Hospital Information, Seoul National University Hospital, Seoul 03080, KoreaOffice of Hospital Information, Seoul National University Hospital, Seoul 03080, KoreaOffice of Hospital Information, Seoul National University Hospital, Seoul 03080, KoreaOffice of Hospital Information, Seoul National University Hospital, Seoul 03080, KoreaOffice of Hospital Information, Seoul National University Hospital, Seoul 03080, KoreaOffice of Hospital Information, Seoul National University Hospital, Seoul 03080, KoreaDepartment of Internal Medicine, Seoul National University Hospital, Seoul 03080, KoreaDepartment of Medical Informatics, The Catholic University of Korea, Seoul 03080, KoreaWith growing interest in machine learning, text standardization is becoming an increasingly important aspect of data pre-processing within biomedical communities. As performances of machine learning algorithms are affected by both the amount and the quality of their training data, effective data standardization is needed to guarantee consistent data integrity. Furthermore, biomedical organizations, depending on their geographical locations or affiliations, rely on different sets of text standardization in practice. To facilitate easier machine learning-related collaborations between these organizations, an effective yet practical text data standardization method is needed. In this paper, we introduce MARIE (a context-aware term mapping method with string matching and embedding vectors), an unsupervised learning-based tool, to find standardized clinical terminologies for queries, such as a hospital’s own codes. By incorporating both string matching methods and term embedding vectors generated by BioBERT (bidirectional encoder representations from transformers for biomedical text mining), it utilizes both structural and contextual information to calculate similarity measures between source and target terms. Compared to previous term mapping methods, MARIE shows improved mapping accuracy. Furthermore, it can be easily expanded to incorporate any string matching or term embedding methods. Without requiring any additional model training, it is not only effective, but also a practical term mapping method for text data standardization and pre-processing.https://www.mdpi.com/2076-3417/10/21/7831text standardizationunsupervised term mappingunsupervised concept normalizationbiomedical text pre-processing |
| spellingShingle | Han Kyul Kim Sae Won Choi Ye Seul Bae Jiin Choi Hyein Kwon Christine P. Lee Hae-Young Lee Taehoon Ko MARIE: A Context-Aware Term Mapping with String Matching and Embedding Vectors Applied Sciences text standardization unsupervised term mapping unsupervised concept normalization biomedical text pre-processing |
| title | MARIE: A Context-Aware Term Mapping with String Matching and Embedding Vectors |
| title_full | MARIE: A Context-Aware Term Mapping with String Matching and Embedding Vectors |
| title_fullStr | MARIE: A Context-Aware Term Mapping with String Matching and Embedding Vectors |
| title_full_unstemmed | MARIE: A Context-Aware Term Mapping with String Matching and Embedding Vectors |
| title_short | MARIE: A Context-Aware Term Mapping with String Matching and Embedding Vectors |
| title_sort | marie a context aware term mapping with string matching and embedding vectors |
| topic | text standardization unsupervised term mapping unsupervised concept normalization biomedical text pre-processing |
| url | https://www.mdpi.com/2076-3417/10/21/7831 |
| work_keys_str_mv | AT hankyulkim marieacontextawaretermmappingwithstringmatchingandembeddingvectors AT saewonchoi marieacontextawaretermmappingwithstringmatchingandembeddingvectors AT yeseulbae marieacontextawaretermmappingwithstringmatchingandembeddingvectors AT jiinchoi marieacontextawaretermmappingwithstringmatchingandembeddingvectors AT hyeinkwon marieacontextawaretermmappingwithstringmatchingandembeddingvectors AT christineplee marieacontextawaretermmappingwithstringmatchingandembeddingvectors AT haeyounglee marieacontextawaretermmappingwithstringmatchingandembeddingvectors AT taehoonko marieacontextawaretermmappingwithstringmatchingandembeddingvectors |