From 40b09af76b74611095e9779da32b1a510b204303 Mon Sep 17 00:00:00 2001
From: Lyudmila Vaseva <vaseva@mi.fu-berlin.de>
Date: Sun, 10 Feb 2019 13:17:11 +0100
Subject: [PATCH] Explore filter actions
---
src/explore.ipynb | 178 +++++++++++++++++++++++++++++++++++++++++++---
1 file changed, 170 insertions(+), 8 deletions(-)
diff --git a/src/explore.ipynb b/src/explore.ipynb
index 4e1fc5d..20476db 100644
--- a/src/explore.ipynb
+++ b/src/explore.ipynb
@@ -4,14 +4,14 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "# An explorative study into EN Wikipedia's edit filter system\n",
+ "# An explorative inquiry into EN Wikipedia's edit filter system\n",
"\n",
"This notebook serves to explore EN Wikipedia's edit filters"
]
},
{
"cell_type": "code",
- "execution_count": 66,
+ "execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
@@ -29,7 +29,7 @@
},
{
"cell_type": "code",
- "execution_count": 62,
+ "execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
@@ -66,7 +66,7 @@
},
{
"cell_type": "code",
- "execution_count": 102,
+ "execution_count": 25,
"metadata": {},
"outputs": [
{
@@ -75,7 +75,8 @@
"text": [
"201\n",
"753\n",
- "600\n"
+ "600\n",
+ "110\n"
]
}
],
@@ -87,7 +88,10 @@
"print (len(df.query('af_enabled==0')))\n",
"\n",
"# Deleted filters\n",
- "print (len(df.query('af_deleted==1')))"
+ "print (len(df.query('af_deleted==1')))\n",
+ "\n",
+ "# Active public filters\n",
+ "print (len(df.query('af_hidden==0 and af_enabled==1')))"
]
},
{
@@ -130,6 +134,165 @@
"print (len(df.query('af_global==0')))"
]
},
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "948\n",
+ "6\n"
+ ]
+ }
+ ],
+ "source": [
+ "# throttled\n",
+ "print (len(df.query('af_throttled==0')))\n",
+ "\n",
+ "print (len(df.query('af_throttled==1')))"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "947\n",
+ " Unnamed: 0 af_id af_hidden af_global af_enabled af_deleted \\\n",
+ "168 168 497 0 0 0 1 \n",
+ "173 173 494 0 0 0 1 \n",
+ "174 174 502 0 0 0 1 \n",
+ "187 187 495 0 0 0 1 \n",
+ "190 190 496 0 0 0 1 \n",
+ "227 227 475 0 0 0 1 \n",
+ "349 349 461 0 0 0 1 \n",
+ "\n",
+ " af_throttled af_group af_timestamp af_actions af_hit_count \\\n",
+ "168 0 feedback 20130108151106 disallow 3660 \n",
+ "173 0 feedback 20130108151035 disallow 3325 \n",
+ "174 0 feedback 20130424011002 disallow 3280 \n",
+ "187 0 feedback 20130108151045 disallow 2697 \n",
+ "190 0 feedback 20130108151054 disallow 2658 \n",
+ "227 0 feedback 20131003210159 NaN 1390 \n",
+ "349 0 feedback 20130411173111 disallow 283 \n",
+ "\n",
+ " af_public_comments manual_tags \\\n",
+ "168 Feedback: Common Vandalism 5 vandalism, harassment? \n",
+ "173 Feedback: Common Vandalism 2 vandalism?, harassment? \n",
+ "174 Feedback: Extremely long words vandalism?, good_faith?, bad_style? \n",
+ "187 Feedback: Common Vandalism 3 vandalism, harassment? \n",
+ "190 Feedback: Common Vandalism 4 vandalism, harassment? \n",
+ "227 Feedback: Vandalism or libel vandalism, harassment \n",
+ "349 Feedback: Vandalism in all caps vandalism, harassment? \n",
+ "\n",
+ " notes \n",
+ "168 deleted; “Merged back into 460. --mlitn†\n",
+ "173 deleted; “Merged back into 460. --mlitn†\n",
+ "174 deleted \n",
+ "187 deleted; “Merged back into 460. --mlitn†\n",
+ "190 deleted; “Merged back into 460. --mlitn†\n",
+ "227 deleted \n",
+ "349 NaN \n"
+ ]
+ }
+ ],
+ "source": [
+ "# group\n",
+ "print (len(df.query('af_group==\"default\"')))\n",
+ "print (df.query('af_group!=\"default\"'))\n",
+ "\n",
+ "# --> so available groups are \"default\" and \"feedback\""
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Helper functions"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 17,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "flatten = lambda x: list(itertools.chain.from_iterable(x))"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Edit filter actions"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 19,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "[('', 413), ('disallow', 406), ('warn', 122), ('tag', 70), ('throttle', 52), ('blockautopromote', 4)]\n"
+ ]
+ }
+ ],
+ "source": [
+ "actions = df['af_actions'].fillna('')\n",
+ "actions_list = [x.split(\",\") for x in list(actions)]\n",
+ "all_actions = flatten(actions_list)\n",
+ "\n",
+ "print(collections.Counter(all_actions).most_common())"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 23,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "[('disallow', 51), ('', 19), ('throttle,disallow', 7), ('throttle', 4), ('tag', 3), ('warn,tag', 2), ('throttle,warn', 2), ('warn', 1), ('disallow,tag', 1), ('warn,disallow', 1)]\n"
+ ]
+ }
+ ],
+ "source": [
+ "# What are the actions of active hidden filters\n",
+ "active_hidden = df.query('af_hidden==1 and af_enabled==1')\n",
+ "print(collections.Counter(list(active_hidden['af_actions'].fillna(''))).most_common())"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 24,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "[('tag', 25), ('warn,tag', 25), ('disallow', 22), ('', 20), ('warn', 12), ('throttle,tag', 2), ('warn,disallow', 2), ('throttle,warn,tag', 1), ('throttle,disallow', 1)]\n"
+ ]
+ }
+ ],
+ "source": [
+ "# What are the actions of active public filters\n",
+ "active_public = df.query('af_hidden==0 and af_enabled==1')\n",
+ "print(collections.Counter(list(active_public['af_actions'].fillna(''))).most_common())"
+ ]
+ },
{
"cell_type": "markdown",
"metadata": {},
@@ -139,7 +302,7 @@
},
{
"cell_type": "code",
- "execution_count": 68,
+ "execution_count": 20,
"metadata": {},
"outputs": [
{
@@ -153,7 +316,6 @@
"source": [
"manual_tags = df['manual_tags']\n",
"manual_tags_list = [x.split(\", \") for x in list(manual_tags)]\n",
- "flatten = lambda x: list(itertools.chain.from_iterable(x))\n",
"all_tags = flatten(manual_tags_list)\n",
"\n",
"print(collections.Counter(all_tags).most_common())"
--
GitLab