CIRCA:Web Interface design(3)


Jump to: navigation, search

These images are a mock up of how the Social Network of the Novel Crowdsourcing tool might work. They make some assumptions:

  • We will be working with small chunks of text like 3 sentences to a paragraph.
  • We will saving information about relationships identified for each chunk and not character location data. In other words each record will be a relationship, entered by a participant, about a chunk, naming two entities, and a form of relation. This means we don't try to save information about character sequences that users may associate with a character. The goal is not to disambiguate pronouns.
  • The system will suggest named entities (characters) based on NER and characters found in passages before.
  • Should the participant want to enter a relation with an entity not presented to them in the left column, they can click the + sign and add a character from a longer list of all the characters found in the novel. In the event that character isn't in that long list then they will have the option to add a new character to the long list.

In this storyboard you see someone choosing the characters, selecting relation settings, and then submitting (adding) to the database. At that point they see the relation added.

They would then go on to do the next passage.

Personal tools