Skip to content

Latest commit

 

History

History
48 lines (32 loc) · 2.18 KB

bandolo.md

File metadata and controls

48 lines (32 loc) · 2.18 KB
Raw
layout title permalink parent
page
Bandolo
/bandolo/
Architecture

Bandolo

Bandolo is an interactive data collection tool. Primarily used to collect data from humans by "chatting" with them. The bot collects data by asking questions determined by a schema which is used to cohorently shape data from different sources and permit Bandolo to be used any kind of data collection.

Example Question Schema

Question {
     source: [__words__, __setences__, __paragraghs__],
     source_fields: ['meaning', 'example', 'alternative'],
}

The above Schema describes the kinds of questions that would be generated from data passed as a collection from any data source via the source key and the data/information that we seek to collect via the source_fields key which accepts an array.

For the example schema shown above we could have a pratical scenario such as;

question_generator = {
    source: ['wahala', 'paddiman', 'hafa', 'where you deh?'],
    source_fields: ['meaning', 'use', 'example', 'alternative']
}

The question generator above would generate question such as :

  1. What is the meaning of paddiman?
  2. Can you make a statement using hafa?
  3. What is alternative way of saying where you deh?
  4. What is a synonym for paddiman?

  • In the first case the meaning source field is using to generate the question for a noun paddiman could equally have been for wahala.

  • In the second case a question is generated using the use source field which could be used to generateda question asking for an example or they "use"/application of something. For example what is the use of .... This bring to mind the notion of keywords which we could introduce into the source fields.

  • In the third and fourth case questions are generated using alternative which could mean alternative phrasing for a setence (case 3) or synonym for a word (case 4).

Difficulties rising from manually using the schema to generate quetions hint that it would be neccesary to language-annotate data passed in to the source. Such language annotation could designate each member of the source collection as a noun,verb, sentence, question or any other type that may be developed.