diff --git a/README.md b/README.md
index 9da0d8e..d42c5ee 100644
--- a/README.md
+++ b/README.md
@@ -2,3 +2,5 @@
 Processed Amazon Review Dataset for Language Generation (Character Level)
 
 Link to the dataset: https://www.kaggle.com/imdeepmind/language-generation-dataset-200m-samples/
+
+Here for preparing the dataset, I've used tiny fraction (only `amazon_reviews_us_Musical_Instruments_v1_00.tsv.gz` file) of the [Amazon Product Review](https://s3.amazonaws.com/amazon-reviews-pds/readme.html) dataset.