https://github.com/idsia/crema-adaptive

Experiments on adaptive surveys using credal and bayesian networks with our CREMA library.
https://github.com/idsia/crema-adaptive

adaptive bayesian credal inference network survey

Last synced: 19 days ago
JSON representation

Experiments on adaptive surveys using credal and bayesian networks with our CREMA library.

• Host: GitHub
• URL: https://github.com/idsia/crema-adaptive
• Owner: IDSIA
• License: lgpl-3.0
• Created: 2020-11-09T12:36:56.000Z (about 4 years ago)
• Default Branch: master
• Last Pushed: 2021-06-09T20:52:12.000Z (over 3 years ago)
• Last Synced: 2024-11-11T14:47:55.715Z (3 months ago)
• Topics: adaptive, bayesian, credal, inference, network, survey
• Language: Java
• Homepage:
• Size: 463 KB
• Stars: 0
• Watchers: 5
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# CREMA - Adaptive experiments

### Description

An application of the the [CREMA](https://github.com/IDSIA/crema) Library.

### Authors

**Giorgia Adorni** - [GiorgiaAuroraAdorni](https://github.com/GiorgiaAuroraAdorni) - [email protected]  

**Alessandro Antonucci** - [alessandroantonucci](https://github.com/alessandroantonucci) - [email protected]  

**Claudio Bonesana** - [cbonesana](https://github.com/cbonesana) - [email protected]  

**Francesca Mangili** - [mangilif](https://github.com/mangilif) - [email protected]

### Usage

In the file `AdaptiveTests` is contained the code to make inference using the *CREMA* libray.

In the files `BNsAdaptiveSurveyTest` and `CNsAdaptiveSurveyTest` are contained examples of usage of the library:

a survey test for a single student is created, each students will have its lists of questions, its personal test,

and its answer sheet. At the end the skills of the students are returned.

The survey is performed using a Bayesian network, in the first case, while a Credal one, in the second.

The files `BNsAdaptiveSurveySimulation` and `CNsAdaptiveSurveySimulation` perform a simulation of a case study in

order to measure the performance of the model. In fact, the only difference from the two previous "Tests" is that the

answers to the questions are sampled using the probabilities of answering correctly or not to a given question. 

The student's real answers are used only as final validation.  

### Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.