-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy patheda.R
34 lines (22 loc) · 1.01 KB
/
eda.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
library(tidyverse)
library(janitor)
library(here)
library(readxl)
df <- read_csv(here("data", "manual", "hand_annotated_flags.csv"))
# Answers to the AI question, according to [our crowdsourced review](https://www.muckrock.com/news/archives/2024/oct/02/help-us-investigate-how-the-federal-government-uses-ai-for-foia/)
# Agencies that were ambiguous or response needed second look because crowdsource response didn't intially provide text from the report
# Text was then added back to these responses
check_text <- df %>%
filter(Flag == 9)
# Agencies that answered no the question
answered_no <- df %>%
filter(Flag == 0)
# Agencies that answered yes or were ambiguous, but upon review of the text, seemed more to reference machine learning ediscovery tools and not AI
not_ai <- df %>%
filter(Flag == 1)
# Agencies that seemed to mention testing or aquiring AI
testing <- df %>%
filter(Flag == 2)
# Agencies that gave a good description of AI beyond ediscovery tools
yes_and <- df %>%
filter(Flag == 3)