ChuGyouk commited on
Commit
3bf7d1d
·
verified ·
1 Parent(s): ae96053

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +24 -6
README.md CHANGED
@@ -22,11 +22,14 @@ I use DeepL for automatic translation and manually reviewed results.
22
 
23
  There are some concerns and warnings for this dataset.
24
 
25
- 1.
26
- 2. I recommend **NOT to use this dataset for Korean medical dialogue summarization task, but to use JUST DIALOGUES for training something like Chatbot**. I did not reviewed for the section_text in detail. More specifically, I've seen cases where "noncontributory" is translated as "비기여" and "negative" is translated as "음수" instead of "부정" or "음성".
27
- 3. There are cases where the translation becomes strange due to some **fundamental differences between English and Korean**. This has not been directly modified during review (for consistency). **For example**, for the doctor's question "And no shortness of breath or wheezing that you've noticed?(숨가쁨이나 쌕쌕거림은 없었나요?)", the response in English is 'No', but it is natural to response as "네" in Korean.
28
- 4. The content may not be helpful due to **cultural differences between the US and Korea**. Simple **example** is the unit of measurements; temperature(°F vs °C), weight(lb vs kg). **Also**, cannabis/marijuana is legal in some states in the US, though it is illegal in Korea.
29
- 5. **Some medical abbreviations are not translated well**. Some have been translated well(ex:COPD - 만성폐쇄성폐질환), but others have been translated incorrectly (ex: CVA - 심근경색) or remain untranslated at all(ex: MCV).
 
 
 
30
 
31
  ## Example
32
 
@@ -70,4 +73,19 @@ There are some concerns and warnings for this dataset.
70
 
71
  ## Reference
72
 
73
- [1] Abacha, A. B., Yim, W. W., Fan, Y., & Lin, T. (2023, May). An empirical study of clinical note generation from doctor-patient encounters. In Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics (pp. 2291-2302).
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
22
 
23
  There are some concerns and warnings for this dataset.
24
 
25
+ 1. I recommend **NOT to use this dataset for Korean medical dialogue summarization task, but to use JUST DIALOGUES for training something like Chatbot**. I did not reviewed for the section_text in detail. More specifically, I've seen cases where "noncontributory" is translated as "비기여" and "negative" is translated as "음수" instead of "부정" or "음성".
26
+
27
+ 2. There are cases where the translation becomes strange due to some **fundamental differences between English and Korean**. This has not been directly modified during review (for consistency). **For example**, for the doctor's question "And no shortness of breath or wheezing that you've noticed?(숨가쁨이나 쌕쌕거림은 없었나요?)", the response in English is 'No', but it is natural to response as "네" in Korean.
28
+
29
+ 3. The content may not be helpful due to **cultural differences between the US and Korea**. Simple **example** is the unit of measurements; temperature(°F vs °C), weight(lb vs kg). **Also**, cannabis/marijuana is legal in some states in the US, though it is illegal in Korea.
30
+
31
+ 4. **Some medical abbreviations are not translated well**. Some have been translated well(ex:COPD - 만성폐쇄성폐질환), but others have been translated incorrectly (ex: CVA - 심근경색) or remain untranslated at all(ex: MCV).
32
+
33
 
34
  ## Example
35
 
 
73
 
74
  ## Reference
75
 
76
+ ```
77
+ @inproceedings{mts-dialog,
78
+ title = {An Empirical Study of Clinical Note Generation from Doctor-Patient Encounters},
79
+ author = "Ben Abacha, Asma and
80
+ Yim, Wen-wai and
81
+ Fan, Yadan and
82
+ Lin, Thomas",
83
+ booktitle = "Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics",
84
+ month = may,
85
+ year = "2023",
86
+ address = "Dubrovnik, Croatia",
87
+ publisher = "Association for Computational Linguistics",
88
+ url = "https://aclanthology.org/2023.eacl-main.168",
89
+ pages = "2291--2302"
90
+ }
91
+ ```