Huaiguo Fu

A Parallel Algorithm to Generate Formal Concepts for Large Data

One of the most effective methods to deal with large data for data analysis and data mining is to develop parallel algorithm. Although Formal concept analysis is an effective tool for data analysis and knowledge discovery, it’s very hard for concept lattice structures to face the complexity of very large data. So we propose a new parallel algorithm based on the NextClosure algorithm to generate formal concepts for large data.

A Comparative Study of FCA-Based Supervised Classification Algorithms

Several FCA-based classification algorithms have been proposed, such as GRAND, LEGAL, GALOIS, RULEARNER, CIBLe, and CLNN & CLNB. These classifiers have been compared to standard classification algorithms such as C4.5, Naive Bayes or IB1. They have never been compared each other in the same platform, except between LEGAL and CIBLe. Here we compare them together both theoretically and experimentally, and also with the standard machine learning algorithm C4.5. Experimental results are discussed.

Partitioning large data to scale up lattice-based algorithm

Concept lattice is an effective tool and platform for data analysis and knowledge discovery such as classification or association rules mining. The lattice algorithm to build formal concepts and concept lattice plays an essential role in the application of concept lattice. We propose a new efficient scalable lattice-based algorithm: ScalingNextClosure to decompose the search space of any huge data in some partitions, and then generate independently concepts (or closed itemsets) in each partition. The experimental results show the efficiency of this algorithm.

Etude et conception d'algorithmes de génération de concepts formels

The lattice algorithm to build formal concepts and concept lattice plays an essential role in the application of concept lattice. In fact, more than ten algorithms for generating concept lattices were published. As real data sets for data mining are very large, concept lattice structure suffers from its complexity issues on such data. The efficiency and performance of concept lattices algorithms are very different from one to another. We need to compare the existing lattice algorithms and develop more efficient algorithm. We implemented and compared the four first algorithms. We analyzed the duality of the lattice-based algorithms. Furthermore, we propose a new efficient scalable lattice-based algorithm: ScalingNextClosure to decompose the search space of any huge data in some partitions, and then generate independently concepts in each partition. The experimental results show the efficiency of this algorithm.

Mining frequent closed itemsets for large data

Mining frequent closed itemsets is one effective method to analyse frequent pattern, and further, to generate association rules. Several algorithms were proposed to generate frequent closed itemsets, including CLOSE, A-CLOSE, CLOSET, CHARM and CLOSET + etc. However its still hard for these algorithms to deal with dense and very large data. In this paper, we analyze the search space of frequent closed itemsets and propose a new decomposition algorithm for mining frequent closed itemsets called PFC. PFC can dynamically generate non-overlapping partitions of the search space and mine frequent closed itemsets in each partition. Furthermore, each partition is independent and only shares the same source data with other partitions. So it is possible to implement PFC with multi-threads or parallel methods, and prune efficiently the search space of frequent closed itemsets. In this study, P FC is implemented in Java. We compare PFC with an authors C++ version of CLOSET + on some large VCI repository datasets and on the worst case. The preliminary experimental results demonstrate good performance of PFC for dealing with dense and very large data.

Clustering binary codes to express the biochemical properties of amino acids

We study four kinds of binary codes of amino acids (AA). Two codes of them are based respectively on biochemical properties, and the two others are generated with artificial intelligence (AI) methods, and are based on protein structures and alignment, and on Dayhoff matrix. In order to give a global significance of each binary code, we use a hierarchical clustering method to generate different clusters of each binary codes of amino acids. Each cluster is examined with biochemical properties to give an explanation on the similarity between amino acids that it contains. To validate our examination, a decision tree based machine learning system is used to characterize the AA clusters obtained with each binary codes. From this experimentation, it comes out that one of the AI based codes allows to obtain clusters that have significant biochemical properties. As a consequence, it appears that even if attributes of binary codes generated with AI methods, do not separately correspond to a biochemical property, they can be significant in the whole. Conversely binary codes based on biochemical properties can be insignificant when forming a whole.