https://github.com/jordicorbilla/ddm
Distributed decision-making system with Jade and Weka
https://github.com/jordicorbilla/ddm
agent artificial-intelligence ddm decision-trees distributed-systems jade java neural-network tree-ensembles weka
Last synced: 9 months ago
JSON representation
Distributed decision-making system with Jade and Weka
- Host: GitHub
- URL: https://github.com/jordicorbilla/ddm
- Owner: JordiCorbilla
- License: other
- Created: 2014-04-19T17:57:55.000Z (about 12 years ago)
- Default Branch: master
- Last Pushed: 2020-04-12T08:47:30.000Z (about 6 years ago)
- Last Synced: 2025-01-31T10:51:10.556Z (over 1 year ago)
- Topics: agent, artificial-intelligence, ddm, decision-trees, distributed-systems, jade, java, neural-network, tree-ensembles, weka
- Language: Java
- Homepage: http://jordicorbilla.github.io/DDM/
- Size: 8.95 MB
- Stars: 8
- Watchers: 5
- Forks: 5
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# DDM
This repository contains Jordi Corbilla's Bsc dissertation: **Distributed decision-making system**, published by Universitat Oberta de Catalunya in 2014 [http://openaccess.uoc.edu/webapps/o2/handle/10609/32761]. The dissertation PDFs and the dissertation sources are licensed under the Creative Commons Attribution license, as described in the LICENSE file.
## Distributed decision-making system with Jade and Weka
[](https://app.box.com/s/q55d26162yrmsviu9259) [](https://app.box.com/s/q55d26162yrmsviu9259) [](https://app.box.com/s/q55d26162yrmsviu9259) [](https://app.box.com/s/q55d26162yrmsviu9259) [](https://app.box.com/s/q55d26162yrmsviu9259) [](https://app.box.com/s/q55d26162yrmsviu9259) [](https://app.box.com/s/q55d26162yrmsviu9259)
DDM is a framework that combines intelligent agents and Artificial Intelligence traditional algorithms such as classifiers. The central idea of this project is to create a multi-agent system that allows to compare different views into a single one.
Two types of agents:
- Manager
- Classifier
Manager Agent processes the dataSets and sends the training data and the data to classify to every classifier instance.
Classifier agent, uses Weka to classify the input data using algorithms such as decision tree (J48), nearest neightbour (IBk) or neuronal networks (MLP) and returns the classification result to the Manager. The manager then creates a decision based on the input from every agent and writes the output with the initial data for comparison purposes.
Diagram below displays the architecture:

#### License
Creative Commons Attribution-NonCommercial 4.0 International Public License