Meeting Notes 20090116

From Nemo

Jump to: navigation, search

1/16/09

Phone meeting (Dejing, Gwen, Bob and Haishan)

Topics

  1. Review issues with TKDD paper draft
  2. Review issues with Haishan's report on replication of case studies from KDD07
  3. Discuss to-do list for everyone 

Gwen pointed out that we should include the report for experiments on LP2 dataset in the TKDD submission. And we should use the same number of patterns across different groups in the same dataset.

Gwen also pointed out that the 'pt#_fac#?' column in the WL3a dataset indicates the auto-labeling result, while the 'pattern_present' column indicates expert-labeling. We should conduct experiments based on sets of pattern factors generated according to these two criteria respectively.

  1. To-do list for ~Jan 13-Feb 1 
    1. Gwen and Bob will comment Haishan's KDD replication report 
    2. Haishan will re-run the experiment on LP1 dataset and use 3 patterns for both group 1 and 2 data. 
    3. Haishan will run the same experiment on LP2 dataset and update the report accordingly 
    4. Haishan will conduct experiments based on WL3a dataset based on the auto-labeling result ('pt#_fac#' column) and update the report accordingly 
    5. Gwen and Bob will provide a set of expert-defined rules for WL3a dataset for Dejing to analyze. 
    6. Dejing will come up a comparison between rules generated from data mining and the above expert-defined rules. 
/wiki/images/1/17/Fish1.png /wiki/images/e/ea/Fish2.png /wiki/images/f/fa/Fish3.png /wiki/images/f/ff/Fish4.png /wiki/images/4/40/Fish5.png /wiki/images/c/c5/Fish6.png
Personal tools