Commit
•
94a1edd
1
Parent(s):
ca3d39b
Update README.md
Browse files
README.md
CHANGED
|
@@ -3,7 +3,7 @@ license: odc-by
|
|
| 3 |
---
|
| 4 |
|
| 5 |
This is a redistribution of the dataset published on Kaggle by MATTIA ZENI, which is a 30 GB json with youtube video statistics from 2013 sampled using wordnet (it's broad).
|
| 6 |
-
This dataset here is parsed into tabular form. The main table is 1millvids_extended.parquet, the primary key is always "X_id".
|
| 7 |
I ran a linear regression on cumulative views for a sorting that reflects virality (variable "views_cummulative_linear_reg_coef") and added a few key stats on the longitudinal numbers, if available.
|
| 8 |
The other tables are long format records of longitudinal youtube engagement data.
|
| 9 |
Note that not all videos are still available.
|
|
|
|
| 3 |
---
|
| 4 |
|
| 5 |
This is a redistribution of the dataset published on Kaggle by MATTIA ZENI, which is a 30 GB json with youtube video statistics from 2013 sampled using wordnet (it's broad).
|
| 6 |
+
This dataset here is parsed into tabular form. The main table is 1millvids_extended.parquet, the primary key is always "X_id" (the unique part of the url).
|
| 7 |
I ran a linear regression on cumulative views for a sorting that reflects virality (variable "views_cummulative_linear_reg_coef") and added a few key stats on the longitudinal numbers, if available.
|
| 8 |
The other tables are long format records of longitudinal youtube engagement data.
|
| 9 |
Note that not all videos are still available.
|