added cleaned dataset

README.md

@@ -1,6 +1,10 @@
 ---
 configs:
 - config_name: default
+  data_files:
+  - split: train
+    path: "stackoverflow/train_clean.csv"
+- config_name: raw
   data_files:
   - split: train
     path: "stackoverflow/train.csv"
@@ -10,8 +14,6 @@ configs:
 
 Working doc: https://docs.google.com/document/d/1h585bH5sYcQW4pkHzqWyQqA4ape2Bq6o1Cya0TkMOQc/edit?usp=sharing
 
-The code to create this dataset is [here](https://github.com/ag2435/arXiv-agent/datasets/stackexchange).
-
 <!-- 1. We run the following query on StackExchange Data Explorer (note that this can be done programmatically using BigQuery for the StackOverflow site, but not the other sites)
 ```
 SELECT *
@@ -41,6 +43,51 @@ FROM answers a
 JOIN questions q ON q.Id = a.ParentId
 ``` -->
 
+BigQuery query (see [so_bigquery.ipynb](so_bigquery.ipynb)):
+```
+CREATE TEMP TABLE answers AS
+SELECT *
+FROM bigquery-public-data.stackoverflow.posts_answers
+WHERE LOWER(Body) LIKE '%arxiv%';
+
+CREATE TEMPORARY TABLE questions AS
+SELECT *
+FROM bigquery-public-data.stackoverflow.posts_questions;
+
+SELECT *
+FROM answers
+JOIN questions ON questions.id = answers.parent_id;
+```
+
+NOTE: BigQuery only has the StackOverflow site data, not the other sites.
+So if we want to query the other sites, we would probably want to download the data dump to a cluster and run a SQL server.
+
+Columns in the raw query output:
+```
+'id', 
+'title', 
+'body', 
+'accepted_answer_id', 
+'answer_count', 
+'comment_count', 
+'community_owned_date', # present only if post is community wiki'd
+'creation_date',
+'favorite_count',
+'last_activity_date', 
+'last_edit_date',
+'last_editor_display_name', 
+'last_editor_user_id',
+'owner_display_name', 
+'owner_user_id',
+'parent_id', # if post is answer, then this is the question id; if post is question, this is None
+'post_type_id', # 1 = QUESTION, 2 = ANSWER
+'score',
+'tags',
+'view_count',
+```
+
+(Official database schema)[https://meta.stackexchange.com/questions/2677/database-schema-documentation-for-the-public-data-dump-and-sede/2678#2678]
+
 ## File structure
 
 Each folder represents the different StackExchange [sites](https://stackexchange.com/sites#oldest) (~182). The largest one is StackOverflow.

stackoverflow/train_clean.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27ca6e8f7c0336abd15ea01858ac44571b0136342eadfbe23678d7b084ab280a
+size 12386618