DMTL (Data Mining Template Library) - A generic C++ based library for mining structured patterns such as sets, sequences, trees and graphs. The library provides implementation of popular frequent pattern mining algorithms.
Be the first to post a text review of Data Mining Template Library. Rate and review a project by clicking thumbs up or thumbs down in the right column.
1.0 release - 07/26/2006 ------------------------ - Added a new back end - file_storage_manager. Now the VATs can be stored in a persistent storage. The file "outfile.dat" is used to house the temporary file cache. IMPORTANT NOTE: The storage manager scales beyond 2 Gigabytes limit only when compiled with gcc 4.0 (this is because in gcc4.0 the fstream class can handle > 2GB files). - Added another set of properties - Tokenizer property. These properties control the tokenizer that would be invoked. For different input file formats new tokenizers can be defined. As an example, a FASTA format tokenizer has been added for biosequences. - Improved the sequence mining implementation of induced and embedded sequences. - Fixed bug in embedded/induced sequence mining which occurred because the gap between consecutive items in a sequence was hard-coded. - Changed the command line inputs. -i, -s and -p are added for input file, support and print patterns, respectively. - Includes support for multi-set pattern. NOTE: Right now the multi-set pattern does not have a file_storage_manager as a back end. - Added LINE_SZ in tokenizer_utils.h so that long input lines can be parsed from the input files. Alpha 0.1 - 11/30/2005 ---------------------- - Initial alpha release.
Be the first person to add a text review.
Copyright © 2009 Geeknet, Inc. All rights reserved. Terms of Use
Thanks for your rating!
Would you also like to write a review?
Thanks for your review!
Get credit for your review by logging in via OpenID. Click your account provider: