projectdescription

projectdescription - CSE532 Artificial Intelligence...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: CSE532 Artificial Intelligence PROJECT DESCRIPTION BAKARY DATA - on the course web page. This is a classification data with TYPE DE ROCHE (Rock Type) as a CLASS attribute. There are 98 records with 48 attributes and 6 classes. Classes are: C1 : R. Carbonatees AND R. Carbonatees impures C2 : Pyrate C3 : Charcopyrite C4 : Galene C5 : Spahlerite C6 : Sediments terrigenes Most important attributes (as determined by the expert) are: S, Zn, Pb, Cu, CaO+MgO, CaO, MgO, Fe2O3 This is a real life experimental data and it contains a lot of missing data (no value). THE PROJECT GOAL is to use an Internet based CLASSIFICATION TOOL to generate sets of DISCRIMINANT RULES describing the content of the data. You can choose one you like, or use WEKA: http://www.cs.waikato.ac.nz/˜ ml/weka/index.html) The project has to follow all steps of Learning Process: Data Preparation that includes attributes selection, cleaning the data, filling the missing values, etc... Data preprocessing : must use at least 2 methods of data discretization, and compare the final results obtained after each of them. Learning Proper : for each experiment describe below use a classification tool for rules generation applied to the TWO sets of preprocessed data and compare the results. Discriminant Rules Generation Experiments ; you have to perform 3 experiments (all on the same preprocessed data) Experiment 1 : use all records to find rules for the full classification; i.e. rules describing all classes C1- C6 simultaneously. Experiment 2 : use all records to find rules contrasting class C1 with all others 1 Experiment 3 : repeat Experiment 1 for all records with the most important attributes only. Write a detailed Project Description with methods, motivations, results and submit it to the Professor in a folder (and CD) on the day of your PROJECT PRESENTATION. Project Presentation : each student, or a group will be given 10-15 minutes to present the project and results. 2 ...
View Full Document

This note was uploaded on 01/25/2012 for the course CSE 352 taught by Professor Wasilewska,a during the Fall '08 term at SUNY Stony Brook.

Page1 / 2

projectdescription - CSE532 Artificial Intelligence...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online