This paper proposes a K-Means clustering method based on genetic algorithm. We compare our method with the traditional K-Means method and clustering method based on simple genetic algorithm. The comparison proves that...This paper proposes a K-Means clustering method based on genetic algorithm. We compare our method with the traditional K-Means method and clustering method based on simple genetic algorithm. The comparison proves that our method achieves a better result than the other two. The drawback of this method is a comparably slower speed in clustering.展开更多
For the prediction of protein structure, we propose the solution based on parallel genetic algorithm. Thissolution has modified the existing algorithm by omitting the process of simulated annealing, while compensating...For the prediction of protein structure, we propose the solution based on parallel genetic algorithm. Thissolution has modified the existing algorithm by omitting the process of simulated annealing, while compensating thismodification by parallel programming to maintain the results quality. Experiments have demonstrated that the newsolution has higher processing speed and same results quality. This solution also underlines the necessity of introduc-ing the idea of parallel programming into the study of bioinformatics.展开更多
This paper proposes a new method of Online Analytical Processing on EMBL Nucleotide SequencesDatabase. This scheme is used to automatically restore flat file data into relational database, which is then convertedinto ...This paper proposes a new method of Online Analytical Processing on EMBL Nucleotide SequencesDatabase. This scheme is used to automatically restore flat file data into relational database, which is then convertedinto OLAP's data marts. Both the quality and speed of analysis will be greatly improved based on the data marts. Webelieve that this method is a powerful and flexible tool and can be seen as successful application of data mining inmolecule Biology.展开更多
文摘This paper proposes a K-Means clustering method based on genetic algorithm. We compare our method with the traditional K-Means method and clustering method based on simple genetic algorithm. The comparison proves that our method achieves a better result than the other two. The drawback of this method is a comparably slower speed in clustering.
文摘For the prediction of protein structure, we propose the solution based on parallel genetic algorithm. Thissolution has modified the existing algorithm by omitting the process of simulated annealing, while compensating thismodification by parallel programming to maintain the results quality. Experiments have demonstrated that the newsolution has higher processing speed and same results quality. This solution also underlines the necessity of introduc-ing the idea of parallel programming into the study of bioinformatics.
文摘This paper proposes a new method of Online Analytical Processing on EMBL Nucleotide SequencesDatabase. This scheme is used to automatically restore flat file data into relational database, which is then convertedinto OLAP's data marts. Both the quality and speed of analysis will be greatly improved based on the data marts. Webelieve that this method is a powerful and flexible tool and can be seen as successful application of data mining inmolecule Biology.
