Table 5 Description of extracted features in the technical validation.
From: K-EmoPhone: A Mobile and Wearable Dataset with In-Situ Emotion, Stress, and Attention Labels
Feature | Feature type | Description |
|---|---|---|
Pre- and post-surveys | ||
PIF#AGE | Num. | The age of a participant. |
PIF#GEN | Cat. | The gender of a participant. |
PIF#BFI_OPN | Num. | The openness score in the BFI questionnaire. |
PIF#BFI_CON | Num. | The conscientiousness score in the BFI questionnaire. |
PIF#BFI_NEU | Num. | The neuroticism score in the BFI questionnaire. |
PIF#BFI_EXT | Num. | The extroversion score in the BFI questionnaire. |
PIF#BFI_AGR | Num. | The agreeableness score in the BFI questionnaire. |
PIF#PSS | Num. | The degree of perceived stress score during the data collection period derived by the PSS questionnaire |
PIF#PHQ | Num. | The degree of depression severity during the data collection period derived by the PHQ questionnaire |
PIF#GHQ | Num. | The degree of psychiatric well-being during the data collection period derived by the GHQ questionnaire |
Pre-processed categorical sensor data (e.g., APP_CAT) | ||
{DATA}#VAL = {VALUE} | Cat. | TRUE if the value recorded at the time nearest to a given timestamp is equals to ‘VALUE’, FALSE otherwise. |
{DATA}#DSC | Num. | The duration between the latest value changes and a given timestamp. |
{DATA}#DSC = {VALUE} | Num. | The duration between the time that a given ‘VALUE’ was recently recorded and a given timestamp. |
{DATA}#ETP#{WINDOW} | Num. | The information entropy of readings within a given time window. |
{DATA}#ASC#{WINDOW} | Num. | The number of changes between consecutive readings within a given time window. |
{DATA}#DUR = {VALUE}#{WINDOW} | Num. | The duration that a ‘VALUE’ lasted within a given time window. |
Pre-processed numerical sensor data (e.g., DAT_RCV) | ||
{DATA}#VAL | Num. | The value recorded at the time nearest to a given timestamp |
{DATA}#AVG#{WINDOW} | Num. | The sample mean of data within a given time window. |
{DATA}#STD#{WINDOW} | Num. | The sample standard deviation of data within a given time window. |
{DATA}#SKW#{WINDOW} | Num. | The sample skewness deviation of data within a given time window. |
{DATA}#KUR#{WINDOW} | Num. | The sample kurtosis deviation of data within a given time window. |
{DATA}#ASC#{WINDOW} | Num. | The sum of absolute differences of data within a given time window. |
{DATA}#BEP#{WINDOW} | Num. | The binned entropy of data within a given time window. |
{DATA}#MED#{WINDOW} | Num. | The median of data within a given time window. |
{DATA}#TSC#{WINDOW} | Num. | The time-series complexity estimate87 of data within a given time window. |
In-situ questionnaires | ||
ESM#DOW = {VALUE} | Cat. | TRUE if the day of the week when a given prompt was triggered equals ‘VALUE’ (which can be either MON: Monday; TUE: Tuesday; WED: Wednesday; THU: Thursday; FRI: Friday; SAT: Saturday or SUN: Sunday), FALSE otherwise. |
ESM#WKD | Cat. | TRUE if the time when a participant received a given prompt is a weekend, FALSE otherwise. |
ESM#HRM = {VALUE} | Cat. | TRUE if the name of the hour when a given prompt was delivered equals ‘VALUE’ (which can be either DAWN: 6AM–9AM; MORNING: 9AM–12PM; AFTERNOON: 12PM–3PM; LATE_AFTERNOON: 3PM–6PM; EVENING: 6PM–9PM; NIGHT: 9PM - 12AM; or MIDNIGHT: 12AM - 6AM), FALSE otherwise. |
ESM#LIK#{WINDOW} | Num. | A prior likelihood of being in a HIGH affective state (i.e., the proportion of HIGH labels over whole labels within a given time window) |
