Over the last decades, a lot of research has been carried out to bring forward many nature-inspired, optimization techniques. The behaviour pattern of natural phenomena such as evolution of species, working of neural networks etc. has been effectively simulated to perform various computing tasks. SitoLIB is an open source library for social impact theory based optimizer (SITO). The goal is to develop an easy to understand, general-purpose software library which can be incorporated in application-specific systems. The present version of library includes the binary version of the optimizer. Our implementation is based on theory of social impact given by [Latane, 1981] and pseudo code of the optimizer given by [Macas, 2008]. So far, three different variants of SITO are implemented in the library for minimization of objection function which includes • OSITO (original SITO algorithm), • SSITOsum (Simplified SITO with SUM rule), SSITOmean Simplified SITO with MEAN rule) and • GSITO (Galam-inspired SITO). These variants have been effectively brought into use in different applications such as feature subset selection using UCI machine learning repository datasets [Macas, 2007], itongue optimization [Bhondekar, 2011], and enhancing e-nose performance [R. Kaur, 2012]. The source code is available from the website sitolib.org. It can be compiled on Microsoft Windows. The usage(Sito Library.pdf) and example usage script(sitodriver.m) is included in the rar file.
- Changes to previous version:
- BibTeX Entry: Download
- URL: Project Homepage
- Supported Operating Systems: Windows
- Data Formats: Matlab
- Tags: Feature Selection, Dimensionality Reduction, Machine Learning, Optimization, Algorithm, Evolutionary Optimization, Human Opinion Formation Based Optimizer, Social Impact Theory
- Archive: download here
No one has posted any comments yet. Perhaps you'd like to be the first?
Leave a comment
You must be logged in to post comments.