Probabilistic Text Change Detection Using an Immune Model
- Resource Type
- Conference
- Authors
- Polla, Matti; Honkela, Timo
- Source
- 2007 International Joint Conference on Neural Networks Neural Networks, 2007. IJCNN 2007. International Joint Conference on. :1109-1114 Aug, 2007
- Subject
- Computing and Processing
Components, Circuits, Devices and Systems
Signal Processing and Analysis
Text analysis
Natural language processing
Neural networks
Immune system
Information analysis
Change detection algorithms
Collaboration
Natural languages
Detectors
Data security
- Language
- ISSN
- 2161-4393
2161-4407
We present a probabilistic approach for detecting and analyzing changes in natural language motivated by biological immune systems. Contrary to traditional methods based on message-digest algorithms and line-by-line comparisons of two files, the proposed algorithm employs an implicit negative representation of text segments in the form of detector strings. A characteristic property of the presented change detection method is that it allows the analysis to be done without revealing the full contents of the original data to the authenticator. Implications of this property to security applications are out-lined and an experiment is conducted to show how several incremental changes to a collaboratively maintained document can be analyzed.