A compressed large language model embedding dataset of ICD 10 CM descriptions

BMC Bioinformatics

Table 2 The autoencoder validation performance ordered by year

Year of Published ICD-10-CM Code	Embedding dimension	Mean square error	Coef. of determination
2019	10	0.593	0.086
2019	50	0.388	0.056
2019	100	0.336	0.049
2019	1000	0.400	0.058
2020	10	0.593	0.086
2020	50	0.388	0.056
2020	100	0.336	0.049
2020	1000	0.400	0.058
2021	10	0.594	0.086
2021	50	0.389	0.056
2021	100	0.337	0.049
2021	1000	0.401	0.058
2022	10	0.595	0.086
2022	50	0.390	0.056
2022	100	0.338	0.049
2022	1000	0.402	0.058

ISSN: 1471-2105