Gene Expression: Theory and Applications in Evolutionary Computation and Distributed Data Mining

本研究项目的目标是通过遵循基因表达的自然过程(DNA-mRNA-蛋白质)的基本机制，开发一种高效的可扩展的进化算法，用于搜索、优化和机器学习。这项工作还应用所开发的算法来解决一个分布式数据挖掘问题。该方法包括：(1)根据概率和近似表示构造来抽象转录(DNA-mRNA)、翻译(mRNA-蛋白质序列)和折叠(序列到蛋白质的3D折叠结构)的操作；(2)利用发展的理解来开发搜索算子；以及(3)将这些机制整合到一个实验性的分布式数据挖掘系统中，用于从分布式数据中发现知识。这项工作的成功完成，将极大地促进进化自适应算法及其应用领域的发展。它将为解决大型优化、机器学习和数据挖掘问题提供一种可伸缩的方法。此外，它还将从进化计算的角度提供对自然基因表达过程的基本理解。Http://www.eecs.wsu.edu/~hillol

The goal of this research project is to develop an efficient scalable evolutionary algorithm for search, optimization, and machine learning by following the underlying mechanism of the natural process of gene expression (DNA - mRNA - Protein). This work also applies the developed algorithms to solve a distributed data mining problem. The approach consists of: (1) abstracting the transcription (DNA - mRNA), translation (mRNA - Protein sequence), and folding (sequence to 3D folded structure of protein) operations in the light of probabilistic and approximate representation construction; (2) development of search operators using the developed understanding; and (3) incorporation of these mechanisms into an experimental distributed data mining system for discovering knowledge from distributed data. A successful completion this work will greatly enhance the field of evolutionary adaptive algorithms and its applications. It will offer a scalable approach for solving large optimization, machine learning, and data mining problems. Moreover, it will offer a fundamental understanding of the natural gene expression process from the perspective of evolutionary computation. http://www.eecs.wsu.edu/~hillol