Downloadable Research Source Codes
This page contains source program implementations of algorithms and systems developed in our Woddlab or used in experiments in the lab. Researchers and developers are encouraged to download and use these source codes with proper credit to the source of the codes and system through citations of journal and conference publications attached to the source codes.
1. C++ program source code implementations of
the PLWAP, WAP, and GSP web log sequential mining algorithms can be downloaded
from: “PLWAP and WAP and GSP
codes”.
2. JAVA program source code implementation of
the WEB LOG cleaner system for generating web log access sequences from real
web log servers can be downloaded from: “Webcleaner code for
cleaning real web logs ”.
3. Our own working version of the IBM Quest
data generator in executable code used to generate our data can be found here “datagenerator” and quick
information on how to generate your own data can be found here “howtogeneratedata”.
Example line for generating data with 200K rows, 30 attributes, 5 maximal
pattern sequence length stored in file 200K.data is given below:
gen seq
-ncust 25 -nitems 0.03 -seq.patlen 5 -fname 200K
Journal paper attached to source programs 1 and 2 above is: C.I. Ezeife and Yi Lu, “Mining Web Log sequential Patterns with Position Coded Pre-Order Linked WAP-tree”, the International Journal of Data Mining and Knowledge Discovery (DMKD), Vol. 10, No. -, pp. 5-38, Kluwer Academic Publishers, June 2005.
Conference paper attached to source programs 1 and
2 above is: C.I.. Ezeife and Yi Lu and Yi Liu, “PLWAP Sequential Mining:
Open Source Code paper”, proceedings of
the Open Source Data
Mining Wrokshop on Frequent Pattern Mining Implementations, in conjunction with
ACM SIGKDD,