您的位置:首页  > 论文页面

包含收益的多时间序列频繁模式挖掘模型

发表时间:2014-01-15  浏览量:1401  下载量:612
全部作者: 王水,王乐
作者单位: 宁波大红鹰学院信息工程学院;大连理工大学电子信息与电气工程学部,计算机科学与技术学院
摘 要: 现有的频繁模式挖掘模型不包含对时间序列时间因素的考虑,也缺乏对模式中可变动“收益”因素的考量。针对此问题,提出多时间序列上包含收益的高效用时序频繁模式挖掘模型;给出宏事务和宏模式的形式定义;以归类函数作为事件“相似”的判据,定义宏模式之间的相似关系。并以股票数据为例,给出简化的归类函数表示。该模型为进一步的研究和计算提供了概念基础。
关 键 词: 数量经济学;多时间序列;归类函数;宏模式;相似宏模式;宏频繁模式
Title: Mining model for frequent high utility temporal patterns on multiple time series
Author: WANG Shui, WANG Le
Organization: College of Information Engineering, Ningbo Dahongying University; School of Computer Science and Technology, Faculty of Electronic Information and Electrical Engineering, Dalian University of Technology
Abstract: The frequent pattern mining model in traditional data mining framework neither take into consideration of the temporal factor of time series, nor can it deal with volatile profit information. To address this issue, we propose a framework for mining high utility frequent temporal patterns on multiple time series, provide the formal definitions of mega transaction and mega pattern and define the similarity relationship of two mega patterns based on classification function. Taking stock trade data as an example, a simplified definition of the classification function is provided. This model provides conceptual basis for related financial researches and calculations.
Key words: quantitative economics; multiple time series; classification function; mega-itemset; similarity of mega patterns; mega frequent pattern
发表期数: 2014年1月第1期
引用格式: 王水,王乐. 包含收益的多时间序列频繁模式挖掘模型[J]. 中国科技论文在线精品论文,2014,7(1):62-69.
 
0 评论数 0
暂无评论
友情链接