毕业设计（论文）_基因表达式编程在数据压缩中的应用.docVIP

本科毕业设计说明书（论文） （201届） 指导教师 学科(专业) 所在学院 提交日期  摘要 各行各业的发展产生了大量的数据，如通话记录，股票数据，气象监测等数据都可以看成数据流。为了挖掘蕴藏在这些数据流中信息的价值，需要一种有效的数据流压缩技术对其进行压缩，以减少这些数据流对存储空间的需求，而传统的压缩技术难以满足这些数据流的存储需求。 基因表达式编程(Gene Expression Programming, GEP)融合了遗传算法和遗传程序设计的优点，编码简单，易于进行遗传修饰操作，具有结构和功能的多样性。在解决复杂的函数发现等问题上，GEP表现出优越的性能，在科学计算和商业应用等领域取得了广泛的应用。 本文利用GEP在函数发现上优越的性能，探索多数据流的压缩算法，提出了一种基于GEP的多数据流压缩算法。通过GEP编码和遗传算子的设计，演化出主、从存储数据流中数据元素之间的函数关系，使得在存储这些数据流时，只需存储主存储数据流和对应的函数关系表达式，从而达到数据压缩的目的。本文所采用算法在保证原数据和重构数据具有相同可用性的前提下，虽然数据精度稍有损失，但获得了较高的压缩比。 通过实验验证，基于GEP的多数据流压缩算法是可行的，当数据流元素存储于数据库时，该算法的压缩比接近，为数据流条数，而当数据流存储于文件时，经过该算法压缩的文件能用其他压缩软件进行二次压缩，从而达到更高的压缩比。 关键词：基因表达式编程，数据流，数据压缩，压缩比Abstract The development of all walks of life gives birth to large amounts of data, such as call records, stock data, meteorological monitoring data and so on, and they are regarded as data streams. In order to mining the valuable information in these data streams, an effective data stream compression technology is badly needed, and this compression technology must be able to reduce the storage space of these data streams significantly. But the traditional compression technologies can’t meet the demands of these data streams. Gene Expression Programming(GEP) incorporates both advantages of Genetic Algorithm and Genetic Programming. GEP is simply coded, easy to run with genetic operators, and it also has the structural and functional diversities. In solving complex problems such as function discovery, GEP shows excellent performance and be widely used in scientific computing and commercial application. Taking advantage of the excellent performance of GEP in function discovery, this paper explores the storage means of multiple data streams, proposes a multiple data streams compression algorithm based on GEP. By coding the GEP and the design of genetic operators, the functional relation between the main data stream and the subordinate data stream is evolved, it makes