Algorithms

subspace fuzzy c-means[toggle]

description

Subspace fuzzy c-means algorithm is a clustering algorithm that assings each data items with its memberships to clusters and in each cluster it assings weight to each attributes. The algorithm has been described in detail in papers:
  1. Krzysztof Siminski, Clustering in Fuzzy Subspaces, [in] Theoretical and Applied Informatics, 2012, volume 24, number 4, pp. 313-326. doi: 10.2478/v10179-012-0019-y
  2. Krzysztof Siminski, Neuro-fuzzy system with weighted attributes, [in] Soft Computing, 2014, volume 18, number 2, pp. 285-297. doi: 10.1007/s00500-013-1057-z
The algorithm has been implemented in the C++ language. The code source is provided below.

download

C++ source code (requires a compiler compatible with at least C++11 standard)

citation

If you use this algorithm, please cite the paper: Krzysztof Simiński, Neuro-fuzzy system with weighted attributes, [in] Soft Computing, 2014, volume 18, number 2, pp. 285-297. doi: 10.1007/s00500-013-1057-z
@ARTICLE{id:Siminski2014Neuro, author = {Krzysztof Siminski}, title = {Neuro-fuzzy system with weighted attributes}, journal = {Soft Computing}, volume = {18}, number = {2}, pages = {285--297}, year = {2014}, doi = {10.1007/s00500-013-1057-z}, abstract = {The paper presents the neuro-fuzzy system with weighted attributes. Its crucial part is the fuzzy rule base composed of fuzzy rules (implications). In each rule the attributes have their own weights. In our system the weights of the attributes are numbers from the interval $[0, 1]$ and they are not global: each fuzzy rule has its own attributes' weights, thus it exists in its own weighted subspace. The theoretical description is accompanied by results of experiments on real life data sets. They show that the neuro-fuzzy system with weighted attributes can elaborate more precise results than the system that does not apply weights to attributes. Assigning weights to attributes can also discover knowledge about importance of attributes and their relations.}, }

fuzzy weighted c-ordered means clustering algorithm[toggle]

description

The proposed Fuzzy Weighted C-Ordered Mean (FWCOM) clustering algorithm elaborates clusters in soft subspaces. In each cluster each attribute is assigned a weight from interval [0, 1]. Each attribute has its individual weight (importance) in each cluster. The proposed algorithm applies the ordering technique to effectively reduce the influence of outliers and noise. The algorithm has been described in detail in the paper:
Krzysztof Siminski, Fuzzy weighted C-ordered means clustering algorithm, [in] Fuzzy Sets and Systems, volume 318, 2017, pages 1-33, doi: 10.1016/j.fss.2017.01.001
The algorithm has been implemented in the C++ language. The code source is provided below.

download

C++ source code (requires a compiler compatible with at least C++11 standard)

citation

If you use this algorithm, please cite the paper: Krzysztof Siminski, Fuzzy weighted C-ordered means clustering algorithm, [in] Fuzzy Sets and Systems, volume 318, 2017, pages 1-33, doi: 10.1016/j.fss.2017.01.001
@article{id:Siminski2017Fuzzy, title = "Fuzzy weighted C-ordered means clustering algorithm ", journal = "Fuzzy Sets and Systems ", volume = {318}, pages = {1-33}, year = "2017", issn = "0165-0114", doi = "http://dx.doi.org/10.1016/j.fss.2017.01.001", url = "http://www.sciencedirect.com/science/article/pii/S0165011417300180", author = "Krzysztof Siminski", abstract = "In real life data sets some attributes may have lower importance or even may be completely noninformative. The subspace clustering algorithms have been proposed to handle this. The soft subspace algorithms are vulnerable to noise and outliers. The paper presents a novel algorithm that handles both various importance of attributes and outliers. The proposed Fuzzy Weighted C-Ordered Mean (FWCOM) clustering algorithm elaborates clusters in soft subspaces. In each cluster each attribute is assigned a weight from interval [0, 1]. Each attribute has its individual weight (importance) in each cluster. The proposed algorithm applies the ordering technique to effectively reduce the influence of outliers and noise. The paper is accompanied by numerical experiments." }

fuzzy c-ordered clustering algorithm for incomplete data (fcominc)[toggle]

description

The algorithm clusters incomplete data into interval type-2 fuzzy clusters. The algorithm assigns typicalities to data item, what makes it robust to outliers and noise. The algorithm has been described in detail in papers:
  1. [to be published]
The algorithm has been implemented in the C++ language. The code source is provided below.

download

C++ source code (requires a compiler compatible with at least C++11 standard)

citation

If you use this algorithm, please cite the paper: [to be published]
[to be published]