Skip to main content

Table 10 Predictive features in SWLS scale. Slang, misspellings and unconventional word forms are shown with an asterisk (*). Errors in lemmatization are enclosed in brackets

From: Predicting subjective well-being in a high-risk sample of Russian mental health app users

Feature type

Feature

Translation/Description

Coefficient

Words

спать_[NOUN]

sleep_VERB

41,086

интим_NOUN

intimacy_NOUN (suggestive of ‘intercourse’)

−44,937

орг_NOUN*

org(aniser)_NOUN

23,978

дропнуть_VERB*

quit_VERB

−64,677

тратиться_VERB

spend_VERB

−24,593

отл_UNKN*

fine_UNKN

34,184

пояснение_NOUN

explanation_NOUN

−22,499

стебать_VERB*

bully_VERB (rude)

−28,898

[вифя]_NOUN*

wifi_NOUN

−48,114

спойлерить_VERB*

spoil_VERB

−48,530

ооохнуть_VERB*

gasp_VERB

−44,864

милый_COMP

nice_COMPARATIVE

56,128

[пиздёжа]_NOUN*

lie_NOUN (rude)

−22,727

обжечь_VERB

burn_VERB

−40,019

Sentiment

Negative_month

negative sentiment in the last month

−29

Activity

AppUsage9-12Ratio

Ratio of phone app usage time between 9 and 12 AM normalized by total app usage time

10

AppUsage0-3Ratio

Ratio of phone app usage time between 0 and 3 AM normalized by total app usage time

−8

AppCats

SOCIAL + COMMUNICATION + DATING_0-3/SOCIAL + COMMUNICATION + DATING

Ratio of time logged in Social + Communication + Dating apps between 0 and 3 AM to total time logged in Social + Communication + Dating apps

11

PHOTOGRAPHY_18-21/18-21

Ratio of time logged in Photography apps between 18 and 21 h PM to total time logged in apps between 18 and 21 h PM

8