Datasets:
Sub-tasks:
multi-class-classification
Languages:
English
Size:
1K<n<10K
Tags:
natural-language-understanding
ideology classification
text classification
natural language processing
License:
EricR401S
commited on
Commit
•
d19bb96
1
Parent(s):
5606a6e
files
Browse files- .gitattributes +1 -0
- README.md +34 -43
- analysis.ipynb +10 -0
.gitattributes
CHANGED
@@ -54,3 +54,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
54 |
*.jpeg filter=lfs diff=lfs merge=lfs -text
|
55 |
*.webp filter=lfs diff=lfs merge=lfs -text
|
56 |
*fm.csv filter=lfs diff=lfs merge=lfs -text
|
|
|
|
54 |
*.jpeg filter=lfs diff=lfs merge=lfs -text
|
55 |
*.webp filter=lfs diff=lfs merge=lfs -text
|
56 |
*fm.csv filter=lfs diff=lfs merge=lfs -text
|
57 |
+
cache filter=lfs diff=lfs merge=lfs -text
|
README.md
CHANGED
@@ -169,20 +169,16 @@ The main usage of this dataset is to study linguistic patterns. Running models a
|
|
169 |
|
170 |
Here is an example analysis notebook showing what can be done with this type of data.
|
171 |
|
172 |
-
Example : [https://colab.research.google.com/drive/
|
173 |
|
174 |
### Direct Use
|
175 |
|
176 |
The suitable use cases are to multi-class classification, word clustering or semantic clustering per different groups, summarization modeling, text parsing, and any other natural language processing task.
|
177 |
|
178 |
-
[More Information Needed]
|
179 |
|
180 |
### Out-of-Scope Use
|
181 |
|
182 |
-
This dataset is not meant to be utilized to demonize or mock certain online communities for the trials in life in which individuals find themselves. If the
|
183 |
-
|
184 |
-
|
185 |
-
[More Information Needed]
|
186 |
|
187 |
## Dataset Structure
|
188 |
|
@@ -194,53 +190,51 @@ Currently, this dataset contains
|
|
194 |
- postid : string
|
195 |
- title of the post: string
|
196 |
- text of the post (where applicable) : string
|
197 |
-
- url (if something was embedded) : string
|
198 |
- score : int32
|
199 |
-
-
|
200 |
-
-
|
|
|
|
|
|
|
|
|
|
|
201 |
|
202 |
-
[More Information Needed]
|
203 |
|
204 |
## Dataset Creation
|
205 |
|
206 |
-
###
|
207 |
|
208 |
-
With the rise of the loneliness epidemic and the radicalization of internet content pitting men and women against each other, it is important to seek understanding
|
209 |
|
210 |
-
|
211 |
|
212 |
-
In
|
213 |
-
|
214 |
-
|
215 |
-
|
216 |
-
|
217 |
|
218 |
-
|
|
|
|
|
|
|
|
|
219 |
|
220 |
### Source Data
|
221 |
|
222 |
-
Each record contains a reddit post,
|
223 |
|
224 |
#### Data Collection and Processing
|
225 |
|
226 |
<!-- This section describes the data collection and processing process such as data selection criteria, filtering and normalization methods, tools and libraries used, etc. -->
|
227 |
|
228 |
-
|
229 |
-
|
230 |
-
However, the plan is to increase the amount of records and leverage the ChatGpt API to summarize the messages into categories. In addition, the dates have to be cleaned a little, in order to add use for researches. I am also not sure if I can retrieve comments per post, further augmenting the data.
|
231 |
-
|
232 |
-
[More Information Needed]
|
233 |
|
234 |
#### Who are the source data producers?
|
235 |
|
236 |
The producers of the data are the various redditors who have participated in these spaces.
|
237 |
|
238 |
-
[More Information Needed]
|
239 |
-
|
240 |
-
### Annotations [optional]
|
241 |
-
|
242 |
-
An annotation that is not part of the collection will be the ChatGPT summarizations (future). The subreddit labels are merely the origins of the posts.
|
243 |
-
|
244 |
#### Annotation process
|
245 |
|
246 |
<!-- This section describes the annotation process such as annotation tools used in the process, the amount of data annotated, annotation guidelines provided to the annotators, interannotator statistics, annotation validation, etc. -->
|
@@ -250,7 +244,7 @@ The origin of the posts are the labels of the records.
|
|
250 |
|
251 |
#### Who are the annotators?
|
252 |
|
253 |
-
|
254 |
|
255 |
|
256 |
#### Personal and Sensitive Information
|
@@ -263,14 +257,11 @@ This dataset contains no personally identifiable information with the exception
|
|
263 |
|
264 |
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
265 |
|
266 |
-
A major caveat is that the pink pill and original red pill groups are shadow banned, impeding their scraping process. This is a flaw I recognize because the original red pill movement, which started in books by authors, propagated itself through its internet (reddit) variant, and it spawned all the other pills.
|
267 |
-
|
268 |
-
Another bias point is that there is more red pill content, as a means to compensate for the ban of the original red pill subreddit.
|
269 |
-
|
270 |
-
As such, I caution researchers to balance their datasets where necessary.
|
271 |
|
|
|
272 |
|
273 |
-
|
274 |
|
275 |
### Recommendations
|
276 |
|
@@ -284,15 +275,15 @@ Users should be made aware of the risks, biases and limitations of the dataset.
|
|
284 |
|
285 |
**BibTeX:**
|
286 |
|
287 |
-
[
|
288 |
|
289 |
**APA:**
|
290 |
|
291 |
-
[
|
292 |
|
293 |
## Glossary [optional]
|
294 |
|
295 |
-
Pill ideologies :
|
296 |
|
297 |
In short, according to archetypical definitions
|
298 |
- the red pill is the emancipation of the masculinity in a feminized age and understanding mating strategies with women.
|
@@ -303,8 +294,8 @@ In short, according to archetypical definitions
|
|
303 |
|
304 |
## Dataset Card Authors [optional]
|
305 |
|
306 |
-
steamcyclone, all the redditors from the subreddits
|
307 |
|
308 |
## Dataset Card Contact
|
309 |
|
310 |
-
-
|
|
|
169 |
|
170 |
Here is an example analysis notebook showing what can be done with this type of data.
|
171 |
|
172 |
+
Example : [https://colab.research.google.com/drive/1ELsp4ccdJgAi6R3FH8e5oj1KNllZmZEz?usp=sharing]
|
173 |
|
174 |
### Direct Use
|
175 |
|
176 |
The suitable use cases are to multi-class classification, word clustering or semantic clustering per different groups, summarization modeling, text parsing, and any other natural language processing task.
|
177 |
|
|
|
178 |
|
179 |
### Out-of-Scope Use
|
180 |
|
181 |
+
This dataset is not meant to be utilized to demonize or mock certain online communities for the trials in life in which individuals find themselves. If the user's motive is to push forward some misandrist or misogynistic agenda, please ignore this dataset and kindly let yourself out the door.
|
|
|
|
|
|
|
182 |
|
183 |
## Dataset Structure
|
184 |
|
|
|
190 |
- postid : string
|
191 |
- title of the post: string
|
192 |
- text of the post (where applicable) : string
|
193 |
+
- url (if something was embedded) : string
|
194 |
- score : int32
|
195 |
+
- date : float64
|
196 |
+
- subreddit_subscribers: int64
|
197 |
+
- num_comments: int64
|
198 |
+
- ups: int64
|
199 |
+
- downs: int64
|
200 |
+
- upvote_ratio : float64
|
201 |
+
- is_video: bool
|
202 |
|
|
|
203 |
|
204 |
## Dataset Creation
|
205 |
|
206 |
+
### Short History of the Ideologies
|
207 |
|
208 |
+
With the rise of the loneliness epidemic and the radicalization of internet content pitting men and women against each other, it is important to seek understanding about the roots of the problem. Depending on whom you ask, you'll get a plethora of answers. Jordan Peterson describes it as some type of post-modernist feminist liberalism problem. The Andrew Tates and other conservative archetypes blame the loss of traditionalism. Others blame dating apps and its selection bias effects. The answer may be a combination of these or somewhere in the middle.
|
209 |
|
210 |
+
More specifically, within each of the major pill ideologies, with the exception of the BlackPill, in the most extremist and mild settings, men blame women to some or large degrees, and women blame men to large degrees. As for the latter, it is very common to witness social media trends of women expressing distaste and dissapointing in men, and this has been ocurring for a few years.
|
211 |
|
212 |
+
As a reaction to this treatment, poor dating outcomes, and poor life outcomes, men and boys from all walks of life sought guidance and self-improvement. In response to this need, the Red Pill was born on the internet, most prominently on Reddit (before being banned), and it specialized in combining informartion from various sources to boost dating outcomes via the understanding of female nature, self-improvement (image and hygiene and career), and social skills. Its main demographic has been lonely men, a unique group of disavowed people who have very little research to understand them. Unfortunately, in recent years, there has been a rise of extremist blue pill ideologies, associated with misandrist speech (women who belittle men), and extremist red pill misogynists (men who belittle women).
|
213 |
+
|
214 |
+
As for Black Pill, it seeks to understand truth through bodies of research. That is their claim.
|
215 |
+
|
216 |
+
It has become quite difficult to isolate less extreme variants of the ideologies from the base variants, and it has also become difficult to sustain academic conversations regarding these topics due to public scrutiny. We have to start somewhere, as can be evidenced by the efforts of all sorts of psychiatrists (Dr. K, Jordan Peterson) and scientists/researchers (Dr. Tali Sharot, Prof. Scott Galloway) around the world.
|
217 |
|
218 |
+
### Curation Rationale
|
219 |
+
|
220 |
+
Now more than ever, polarization is a topic that has gone beyond politics and is now deeply embedded in dating dynamics(which have also become proxies for politics - conservative/liberal dynamics). To make matters worse, especially in the case of male spaces, as substantiated by research and media coverage in recent years, have only been able to exist on the internet due to scrutiny and silencing of male voices, and counter-spaces have emerged to challenge the views held in the differing ideologies. The same extends to the other groups, where speaking publicly on such matters earns weird looks at best and public shame and social exile at worst. In the current social climate, the dominant ideology is most commonly labeled as mild blue pill, occassionally with a tinge of Black Pill.
|
221 |
+
|
222 |
+
In contrast, works of Dr. Alok Kanojia (Dr.K, Healthy Gamer Foundation), serve as a basis to understand the individual behind the pain and help said individual build human connections worth having. To that end, what better way to understand people than to listen to them directly, on a platform's subreddits that were created solely for them to share their thoughts, unfiltered thanks to the anonymity. Can we derive some understanding over the multiple disenfranchised groups from this dataset? Can such understanding be published to ultimately help people become better people, sons/daughters, spouses and partners.
|
223 |
|
224 |
### Source Data
|
225 |
|
226 |
+
Each record contains a reddit post, a couple hundred per subreddit, and has a key title and a post with words to display the intended message by the author. The authors will remain anonymous, as they do not deserve persecution for their thoughts, whether you disagree with them or not.
|
227 |
|
228 |
#### Data Collection and Processing
|
229 |
|
230 |
<!-- This section describes the data collection and processing process such as data selection criteria, filtering and normalization methods, tools and libraries used, etc. -->
|
231 |
|
232 |
+
The empty text fields almost always corresponded to videos, so they have been replaced by empty strings. The curation of the content can be expanded in the future, but for now, over 7,000 records have been curated.
|
|
|
|
|
|
|
|
|
233 |
|
234 |
#### Who are the source data producers?
|
235 |
|
236 |
The producers of the data are the various redditors who have participated in these spaces.
|
237 |
|
|
|
|
|
|
|
|
|
|
|
|
|
238 |
#### Annotation process
|
239 |
|
240 |
<!-- This section describes the annotation process such as annotation tools used in the process, the amount of data annotated, annotation guidelines provided to the annotators, interannotator statistics, annotation validation, etc. -->
|
|
|
244 |
|
245 |
#### Who are the annotators?
|
246 |
|
247 |
+
The subreddit origin and the post authors (by deciding to publish on the specific subreddit) are the label annotators.
|
248 |
|
249 |
|
250 |
#### Personal and Sensitive Information
|
|
|
257 |
|
258 |
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
259 |
|
260 |
+
A major caveat is that the pink pill and original red pill groups are shadow banned, impeding their scraping process. This is a flaw I recognize because the original red pill movement, which started in books by authors, propagated itself through its internet (reddit) variant, and it spawned all the other pills. In other words, the biggest sources of information are locked away, and we have to make use of their closest proxies and/or sibling subreddits.
|
|
|
|
|
|
|
|
|
261 |
|
262 |
+
Another bias point is that there are more red pill groupings, as a means to compensate for the ban of the original red pill subreddit.
|
263 |
|
264 |
+
As such, I caution researchers to balance their datasets where necessary. The next step for this dataset is to expand to take the original Red and Pink Pill subreddits.
|
265 |
|
266 |
### Recommendations
|
267 |
|
|
|
275 |
|
276 |
**BibTeX:**
|
277 |
|
278 |
+
[Blog Post Coming Soon]
|
279 |
|
280 |
**APA:**
|
281 |
|
282 |
+
[Blog Post Coming Soon]
|
283 |
|
284 |
## Glossary [optional]
|
285 |
|
286 |
+
Quick Definitions of the Pill ideologies :
|
287 |
|
288 |
In short, according to archetypical definitions
|
289 |
- the red pill is the emancipation of the masculinity in a feminized age and understanding mating strategies with women.
|
|
|
294 |
|
295 |
## Dataset Card Authors [optional]
|
296 |
|
297 |
+
steamcyclone, all the redditors from the subreddits (anonymized).
|
298 |
|
299 |
## Dataset Card Contact
|
300 |
|
301 |
+
- Look me up.
|
analysis.ipynb
CHANGED
@@ -1,5 +1,10 @@
|
|
1 |
{
|
2 |
"cells": [
|
|
|
|
|
|
|
|
|
|
|
3 |
{
|
4 |
"cell_type": "code",
|
5 |
"execution_count": 1,
|
@@ -24,6 +29,11 @@
|
|
24 |
"test = datasets.load_dataset(\"steamcyclone/Pill_Ideologies-Post_Titles\", trust_remote_code=True)"
|
25 |
]
|
26 |
},
|
|
|
|
|
|
|
|
|
|
|
27 |
{
|
28 |
"cell_type": "code",
|
29 |
"execution_count": null,
|
|
|
1 |
{
|
2 |
"cells": [
|
3 |
+
{
|
4 |
+
"cell_type": "markdown",
|
5 |
+
"metadata": {},
|
6 |
+
"source": []
|
7 |
+
},
|
8 |
{
|
9 |
"cell_type": "code",
|
10 |
"execution_count": 1,
|
|
|
29 |
"test = datasets.load_dataset(\"steamcyclone/Pill_Ideologies-Post_Titles\", trust_remote_code=True)"
|
30 |
]
|
31 |
},
|
32 |
+
{
|
33 |
+
"cell_type": "markdown",
|
34 |
+
"metadata": {},
|
35 |
+
"source": []
|
36 |
+
},
|
37 |
{
|
38 |
"cell_type": "code",
|
39 |
"execution_count": null,
|