Under construction

This is, in it's current form, a project I am using for practice. I am not familiar with Flutter nor Dart before this project, and am experimenting with NLP integration.

Actions

1. Up = good
2. Down = bad

Action attributes

1. level
2. what_happened

• 200 chars max

3. talk_about_it

• full screen
• no char max

4. timestamp
5. device
6. tokenized_what_happened

• a list of tokens relating to this action which can be indexed

7. type

Action levels

1. Little
2. "" = standard
3. Big
4. Big big

Action triggers

1. All actions => "Wanna talk about it?" => talk_about_it = "" || input || "Wanna really talk about it?" => input
2. All Down || Big Up || Big Big Up => "Wanna talk about it?" is bigger
3. Big || Big big => "Wanna talk about it?" requires confirm prompt to supply ""

Action UI integrations

1. autocomplete suggestions based on previous tokenized_what_happened
2. timestamp is close to now? => suggest previous actions from similar timestamps
3. suggestions from similar times or autocomplete are available? => allow "negating" an up or down that was not done this time

Action insights

1. Total ups and downs with adjustable time slider

• default is "2 weeks"
• determines times when clusters of ups or downs began or ended and suggests timeframes for "clean slates"
• visual relations between the attributes of the actions in aggregate (histograms, timelines, graphs, etc), with groupings by timestamp, tokenized_what_happened, level, and/or type.