Gautam Garai

A novel genetic algorithm for automatic clustering

In this paper we have presented a new genetically guided algorithm for solving the clustering problem. The proposed Genetic Clustering Algorithm is basically a two-phase process. At the first phase the original data set is decomposed into a number of fragmented clusters in order to spread the GA search process at the latter phase over the entire space. At the second phase Hierarchical Cluster Merging Algorithm (HCMA) is used. The HCMA is an iterative genetic algorithm based approach that combines some of the fragmented clusters into complete k-cluster. The algorithm contains another component called Adjacent Cluster Checking Algorithm (ACCA). This technique is used for testing adjacency of two segmented clusters so that they can be merged into one cluster. The performance of the algorithm has been demonstrated on several data sets consisting of multiple clusters and it is compared with some well-known clustering methods.

A distributed hierarchical genetic algorithm for efficient optimization and pattern matching

In this paper we propose a new approach in genetic algorithm called distributed hierarchical genetic algorithm (DHGA) for optimization and pattern matching. It is eventually a hybrid technique combining the advantages of both distributed and hierarchical processes in exploring the search space. The search is initially distributed over the space and then in each subspace the algorithm works in a hierarchical way. The entire space is essentially partitioned into a number of subspaces depending on the dimensionality of the space. This is done in order to spread the search process more evenly over the whole space. In each subspace the genetic algorithm is employed for searching and the search process advances from one hypercube to a neighboring hypercube hierarchically depending on the convergence status of the population and the solution obtained so far. The dimension of the hypercube and the resolution of the search space are altered with iterations. Thus the search process passes through variable resolution (coarse-to-fine) search space. Both analytical and empirical studies have been carried out to evaluate the performance between DHGA and distributed conventional GA (DCGA) for different function optimization problems. Further, the performance of the algorithms is demonstrated on problems like pattern matching and object matching with edge map.

A split and merge procedure for polygonal border detection of dot pattern

An approach to find the polygonal border of a dot pattern is proposed. The procedure starts with a convex hull of the dot pattern and obtains the final border by the process of splitting followed by merging. During splitting one or more sides of the convex hull are deleted and new sides are added to take care of the inherent concavity. To obtain a smooth polygonal border, two or more sides are merged into a single one. An advantage of the procedure is that the user can set a priori the number of sides of the polygon. Also, it works quite well if there is a gradual transition of dot density in the pattern. Given the convex hull, the procedure can be executed in O(nm) time for a pattern consisting of n dots and an m-sided polygonal border. Moreover, the procedure can be implemented in a parallel system as the operations on each side of the convex hull polygon are independent as well as localised.

A cascaded genetic algorithm for efficient optimization and pattern matching

Abstract A modified genetic algorithm (GA) based search strategy is presented here that is computationally more efficient than the conventional GA. Here the idea is to start a GA with the chromosomes of small length. Such chromosomes represent possible solutions with coarse resolution. A finite space around the position of solution in the first stage is subject to the GA at the second stage. Since this space is smaller than the original search space, chromosomes of same length now represent finer resolution. In this way, the search progresses from coarse to fine solution in a cascaded manner. Since chromosomes of small length are used at each stage, the overall approach becomes computationally more efficient than a single stage algorithm with the same degree of final resolution. The effectiveness of the proposed GA has been demonstrated for the optimization of some synthetic functions and on pattern recognition problem namely dot pattern matching and object matching with edge map.

A novel hybrid genetic algorithm with Tabu search for optimizing multi-dimensional functions and point pattern recognition

Hybrid evolutionary algorithms are drawing significant attention in recent time for solving numerous real world problems. This paper presents a new hybrid evolutionary approach for optimizing mathematical functions and Point Pattern Recognition (PPR) problems. The proposed method combines a global search genetic algorithm in a course-to-fine resolution space with a local (Tabu) search algorithm. Such hybridization enhances the power of the search technique by virtue of inducing hill climbing and fast searching capabilities of Tabu search process. The approach can reach the global or near-global optimum for the functions in high dimensional space. Tests have been successfully made on several benchmark functions in up-to 100 dimensions. The performance of the proposed algorithm has been compared with other relevant algorithms using non-parametric statistical approaches like Friedman test, multiple sign-test and contrast estimation. Also, the hybrid method with grid based PPR technique has been applied for solving dot pattern shape matching and object matching represented as edge maps. The performance of proposed method compares favorably with relevant approaches reported in the article.

A cascaded pairwise biomolecular sequence alignment technique using evolutionary algorithm

In computational biology, biological sequence alignment is an important and challenging task for sequence analysis. In this paper, we propose a new sequence alignment technique based on a genetic algorithm (GA) for determining the optimal alignment score for a pair of sequences that could be either DNA or protein sequences. The search space requirement of the proposed genetic-based method, named Cascaded Pairwise Alignment with Genetic Algorithm (CPAGA), is reduced by breaking a large space into smaller subspaces. This is performed by decomposing the sequence pair into multiple segments before starting the alignment procedure. Such decomposition enhances the ability of the search process to reach the global or a near-global optimal solution even for the longer sequences. The method was tested using several DNA/protein sequence pairs. We also compared the alignment score of the CPAGA with that of some well-known and relevant alignment techniques. The performance of the CPAGA method and other relevant techniques was assessed by a set of non-parametric statistical approaches, which suggest a superior performance of CPAGA over the other alignment procedures.

A review on multiple sequence alignment from the perspective of genetic algorithm

Sequence alignment is an active research area in the field of bioinformatics. It is also a crucial task as it guides many other tasks like phylogenetic analysis, function, and/or structure prediction of biological macromolecules like DNA, RNA, and Protein. Proteins are the building blocks of every living organism. Although protein alignment problem has been studied for several decades, unfortunately, every available method produces alignment results differently for a single alignment problem. Multiple sequence alignment is characterized as a very high computational complex problem. Many stochastic methods, therefore, are considered for improving the accuracy of alignment. Among them, many researchers frequently use Genetic Algorithm. In this study, we have shown different types of the method applied in alignment and the recent trends in the multiobjective genetic algorithm for solving multiple sequence alignment. Many recent studies have demonstrated considerable progress in finding the alignment accuracy.

An efficient Ant Colony Optimization algorithm for function optimization

In this article we have proposed an efficient Ant Colony Optimization method, namely Guided Ant Colony Optimization (GACO) technique for optimizing mathematical functions. The search process of the optimization approach is directed towards a region or a hypercube in a multidimensional space where the amount of pheromone deposited is maximum after a predefined number of iterations. The entire search area is initially divided into 2n number of hypercubic quadrants where n is the dimension of the search space. Then the pheromone level of each quadrant is measured. Now, the search jumps to the region (new search area) of maximum pheromone level and restarts the search process in the new region. However, the search area of the new region is reduced compared to the previous search area. Thus, the search advances and jumps to a new search space (with a reduced search area) in several stages until the algorithm is terminated. The GACO technique has been tested on a set of mathematical functions with number of dimensions upto 100 and compared with several relevant optimizing approaches to evaluate the performance of the algorithm. It is observed that the proposed technique performs better or similar to the performance of other optimization methods.

A hierarchical genetic algorithm with search space partitioning scheme

A new search technique of genetic algorithm (GA) called hierarchical genetic algorithm (HGA) has been proposed for optimizing various functions in R/sup n/ space. Initially the entire search space is partitioned into a number of subspaces depending on the dimensionality of the search space. The HGA processes are then distributed. The algorithm thus independently runs in each subspace with the advancement of the search from one hypercube to a neighboring hypercube surrounding the current best individual depending on the convergence status of the population and the solution obtained so far in the same subspace. The search process passes through variable resolution (coarse-to-fine) search space as the hypercube dimension is modified with the shift of the search to the neighboring hypercube. The performance of HGA and conventional GA (CGA) has been evaluated for different function optimization problems.

An optimized approach for annotation of large eukaryotic genomic sequences using genetic algorithm

BackgroundDetection of important functional and/or structural elements and identification of their positions in a large eukaryotic genomic sequence are an active research area. Gene is an important functional and structural unit of DNA. The computation of gene prediction is, therefore, very essential for detailed genome annotation.ResultsIn this paper, we propose a new gene prediction technique based on Genetic Algorithm (GA) to determine the optimal positions of exons of a gene in a chromosome or genome. The correct identification of the coding and non-coding regions is difficult and computationally demanding. The proposed genetic-based method, named Gene Prediction with Genetic Algorithm (GPGA), reduces this problem by searching only one exon at a time instead of all exons along with its introns. This representation carries a significant advantage in that it breaks the entire gene-finding problem into a number of smaller sub-problems, thereby reducing the computational complexity. We tested the performance of the GPGA with existing benchmark datasets and compared the results with well-known and relevant techniques. The comparison shows the better or comparable performance of the proposed method. We also used GPGA for annotating the human chromosome 21 (HS21) using cross-species comparisons with the mouse orthologs.ConclusionIt was noted that the GPGA predicted true genes with better accuracy than other well-known approaches.