Design and implementation of incremental algorithm for creation of generalized one-sided concept lattices

2012 IEEE 13th International Symposium on Computational Intelligence and Informatics (CINTI)

2012

Self Cite

In this paper we provide experimental comparison on time complexity of algorithm for creation of Generalized OneSided Concept Lattices according to the sparseness of the input data table between standard and sparse-based implementation. It is an incremental algorithm related to FCA (Formal Concept Analysis), which is ready to be used with various attribute types and to create so-called generalized one-sided concept lattice. While these algorithms are generally exponential, in practice the complexity can be considerably reduced, e.g., for sparse input data tables. We describe sparse-based implementation of two crucial operations in our algorithm and provide experiments with different sparse data tables, where time complexity is studied for comparison of standard and sparse-based implementation of the algorithm.

“…Some details on algorithm and implementation of this algorithm, together with some examples of attributes modeling and output (resulted) concept lattices is possible to found in [5].…”

Section: Gosclmentioning

confidence: 99%

Comparison of standard and sparse-based implementation of GOSCL algorithm

2012 IEEE 13th International Symposium on Computational Intelligence and Informatics (CINTI)

2012

Self Cite

“…Theoretical introduction (with the proofs and basic view of the algorithms) of the model can be found in [7] (for more implementation details and some real data example see [8]). Here we recall only basic information needed for the rest of this paper.…”

Section: Generalized One-sided Concept Latticesmentioning

confidence: 99%

“…The presented approaches works only with one type of attribute (i.e., truth degrees structure for the attributes) which is used for input data table. In one of our papers (see [7]) we have theoretically introduced the generalized approach (in [8] it is possible to find also some real data examples for such model) called Generalized One-Sided Concept Lattice (GOSCL), which is able to process data tables containing also different attribute types like quantitative (with values from interval of reals), ordinal (scale-based), nominal (particular values from some domain), or simply also binary attributes.…”

Section: Introductionmentioning

confidence: 99%

Bisection-based merging algorithm for creation of one-sided concept lattices

2014 IEEE 12th International Symposium on Applied Machine Intelligence and Informatics (SAMI)

2014

Self Cite

In this paper we provide the new version of algorithm for creation model called Generalized One-Sided Concept Lattice (GOSCL). This model provides the specific fuzzy version of data analytical method based on the approach known as Formal Concept Analysis (FCA), which supports data tables containing the multiple types of attributes defined as fuzzy sets. The acquisition of the FCA models is computationally complex task and it is important to find more effective algorithms for their creation. Therefore, we have designed the algorithm for the reduction of the computation times, which is based on the simple division of input data table using bisection-based approach and then merging procedure compose the local models into one finally merged concept lattice for the complete input data. We present the illustrative experiments which prove the applicability of the presented algorithm for sparse data inputs, where it is possible to get significant decrease of computation times. More effective algorithm for sparse data can be useful for the application of FCA-based models in sparse domains like information retrieval or text analysis.

“…Detailed description of the presented algorithm and its implementation in Java, examples of attributes definitions and concept lattices on some real data can be found in [7].…”

Section: Cr)mentioning

confidence: 99%

“…It means that it is possible to combine different truth value structures for attributes, e.g., qualitative attributes with possible values 0 and 1, quantitative attributes from some real-valued interval, ordinal attributes, etc. In paper [7] the details regarding design and implementation of incremental algorithm for creation of generalized one-sided concept lattices (GOSCL) were presented.…”

Section: Introductionmentioning

confidence: 99%

Experimental study on time complexity of GOSCL algorithm for sparse data tables

2012 7th IEEE International Symposium on Applied Computational Intelligence and Informatics (SACI)

2012

Self Cite

In this paper we provide experimental study on time complexity of GOSCL algorithm according to the sparseness of the input data table. GOSCL is incremental algorithm for the creation of Generalized One-Sided Concept Lattices, which is related to well-known Formal Concept Analysis area, but with the possibility to work with different types of attributes and to produce one-sided concept lattice from the generalized one-sided formal context. Generally, FCA-based algorithms are exponential. However, in practice there are many inputs for which the complexity is reduced. One of the special cases is related to the high number of ''zeros'' (bottom elements) in data table for so-called sparse data matrices, which is characteristic for some inputs like document-term matrix in text-mining analysis. We describe experimentally the influence of sparseness of data tables on time complexity of GOSCL with different distributions of zeros generated artificially randomly or according to the standard text-mining datasets.