S-EM: Text Classification Using Positive and Unlabeled Data

S-EM Download Page

Please download our new system, LPU, which includes S-EM and also some newer techniques

S-EM is a text learning or classification system that learns from a set of positive and unlabeled examples (no negative examples). It is based on a "spy" technique, naive Bayes and EM algorithm. The detailed algorithm is given in (Liu, Lee, Yu & Li, ICML'02).

Download and Install

Currently, we only provide the executable (.exe) version of the system (without source) for Windows PC. If you encounter any problem in running the program, please email us. If you MUST have a UNIX version, please let us know also.

1. Download the S-EM system here (Windows PC version).
2. Readme file (explain how to run the system).
3. Extract the files in the zip file to a directory. In this directory, the S-EM directory will be created, which contains 4 files.
   • s-em.exe: The S-EM executable program.
   • An example dataset with 3 files: demo.pos (positive data), demo.unlabel (unlabeled data), demo.test (test data). You can run the dataset (see Readme file on how to run).

If you have downloaded S-EM, please send us an email so that we can put you in our mailinglist to inform you any new versions and bug-fixes. We will also be very happy to hear your comments about S-EM.


free hit counter

Created on Dec 31 2002 by Bing Liu; and Xiaoli Li.