Back

Sampling Statistical Errors in Big Data Research: 3 Cases of Breast Cancer Research

Cho, H.-J.; Jeong, E. S.; Lee,

2021-11-01 oncology
10.1101/2021.10.07.21264601 medRxiv
Show abstract

Breast cancer is a major cause of female death, and various big data analysis methods have been applied to breast cancer. This study lists cases in which big data analysis was applied to breast cancer research. In addition, statistics and percentages from each specific sample were proposed. However, research on the use of big data has a blind spot that relies on sample characteristics. Therefore, before sampling big data, statistical inference should be discussed more precisely through pre-examination and sample statistical errors should be reduced by professional statistical evaluation of the analysis method. In particular, the control and experimental groups should be statistically equivalent.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.