Table 3.
Some of the main context variables with their initial value and a brief description
| Name | Init value | Description |
|---|---|---|
| intent | none | Used for representing the user’s intents. Values are set by the NLU module |
| n | none | A numeric variable used for recognizing correct/incorrect answers. Values are set by the NLU module |
| sentiment | 0 | A numeric variable ranging from –1 to 1 representing the sentiment, from negative to positive, of the last utterance from the user. Values are set by the the NLU module |
| extraversion | none | A numeric variable ranging from 0 to 1 representing the extraversion of a user. Values are set by the personality insight module |
| confidence | none | A numeric variable ranging from 0 to 1 representing the confidence of the system in recognizing a user’s utterance or the user’s intent. Values are set by the speech-to-text module and by the NLU module |
| eyes | normal | Used for representing the current state of the eyes. Values are set by executing actions |
| node | none | Used for representing the current state-transition node of the exercise. Values are set by executing actions |
| num_errors | 0 | Used for representing the current performance of the user in terms of the number of errors made. Values are set by executing actions |
| audio_volume | normal | Used for adapting the audio volume for persons with hearing impairments. Values are set by KOaLa through motivation |
| exercise | none | Used for describing the current rehabilitation excercise. Values are set by KOaLa through motivation |