计算机科学 ›› 2021, Vol. 48 ›› Issue (8): 86-90.doi: 10.11896/jsjkx.210600031

• 数据库&大数据&数据科学* • 上一篇    下一篇

基于FP-Growth算法和GRNN的电力知识文本挖掘

白勇1, 张占龙2, 熊隽迪1   

  1. 1 重庆电力高等专科学校 重庆400053
    2 重庆大学 重庆400030
  • 收稿日期:2021-06-01 修回日期:2021-06-28 发布日期:2021-08-10
  • 通讯作者: 白勇(391990750@qq.com)
  • 基金资助:
    国家自然科学基金(52007011);重庆市教委科学技术重点研究项目(KJZD-K202002601)

Power Knowledge Text Mining Based on FP-Growth Algorithm and GRNN

BAI Yong1, ZHANG Zhan-long2, XIONG Jun-di1   

  1. 1 Chongqing Electric Power Specialist University,Chongqing 400053,China;
    2 Chongqing University,Chongqing 400030,China
  • Received:2021-06-01 Revised:2021-06-28 Published:2021-08-10
  • About author:BAI Yong,born in 1973,master,asso-ciate professor.His main research in-terests include electric energy measurement information processing and data mining.
  • Supported by:
    National Science Foundation of China(52007011) and Chongqing Municipal Education Commission of Science and Technology Research Key Project(KJZD-K202002601).

摘要: 为了提高电力知识文本挖掘的性能,采用FP-Growth算法对影响电力需求的强关联因素进行挖掘,运用广义回归神经网络(General Regression Neural Network,GRNN)算法实现电力需求预测。首先,对待挖掘的电力文本进行指标提取并编码,生成电力文本初始FP-Tree;接着采用FP-Growth算法遍历所有FP-Tree,生成频繁集,过滤掉小于最小支持度的项,留下频数较高的频繁项;然后根据更新后的FP-Tree统计关联项,选择与总用电量增长率关联强的变量生成训练样本;最后采用GRNN算法对电力需求文本进行训练,输入电力需求预测样本,设置平滑因子,通过模式层的输出和加权求和来获得电力需求预测结果。实验结果证明,通过合理设置最小支持度和GRNN的平滑因子,能够获得较好的电力文本挖掘性能,与常用挖掘算法相比,所提算法能够获得更高的电力需求预测准确率。

关键词: FP-Growth算法, 电力文本挖掘, 广义回归神经网络, 频繁集, 平滑因子

Abstract: In order to improve the performance of power knowledge text mining,FP-Growth algorithm is used to mine the strong correlation factors that affect the power demand,and GRNN algorithm is used to realize the power demand forecasting.Firstly,the index of the power text to be mined is extracted and encoded to generate the initial FP-Tree.Then,FP-Growth algorithm traverses all FP-Tree generated frequent sets,filters out the items less than the minimum support,leaves the frequent items with higher frequency.And then according to the updated FP-Tree statistical correlation items,it selects variables with strong correlation with the growth rate of total electricity consumption to generate training samples.Finally,the GRNN algorithm is used to train the power demand text,input the power demand forecasting samples,set the smoothing factor,and obtain the power demand forecasting results through the output and weighted sum of the mode layer.Experimental results show that better power text mining performance can be obtained by setting the minimum support and the smoothing factor of GRNN.Compared with common mining algorithms,this algorithm can obtain higher accuracy of power demand forecasting.

Key words: FP-Growth algorithm, Frequent set, Generalized regression neural network, Power text mining, Smoothing factor

中图分类号: 

  • TP391.1
[1]BAI K F,YANG B,WEI J.Review of power text mining techno-logy[J].Electronic Technology and Software Engineering,2019,168(22):143-144.
[2]ZHANG P M,GU J,ZHANG Q.Model Prediction for Grid-connected Inverter Control Strategy under Weak Power Grid[J].Journal of Chongqing Technology and Business University(Na-tural Science Edition),2019,36(6):100-105.
[3]GUO B J,BIAN X X.Application of Improved Genetic Algo-rithms to Distribution Network Fault Location with DG[J].Journal of Chongqing Technology and Business University(Na-tural Science Edition),2019,36(5):24-30.
[4]KONG C X,YANG A,ZHU L J.Traveling Wave Fault Location of Double-ended Transmission Lines Based on Wavelet Transform[J].Journal of Chongqing Technology and Business University(Natural Science Edition),2019,36(5):31-36.
[5]LIU Y H,HUANG S W,MEI S W,et al.Analysis method of cascading failure mode in power system based on sequential pattern mining[J].Power System Automation,2019,43(6):34-40.
[6]WANG C Y,JIANG Q Y,TANG Y J,et al.Power dispatching fault diagnosis based on text mining of alarm signal [J].Power Automation Equipment,2019,39(4):126-132.
[7]GLIGOR A,VLASA I,DUMITRU C D,et al.Power DemandForecast for Optimization of the Distribution Costs[J].Procedia Manufacturing,2020,46:384-390.
[8]JR A,MKK B.Map-optimize-reduce:CAN tree assisted FP-growth algorithm for clusters based FP mining on Hadoop[J].Future Generation Computer Systems,2020,103:111-122.
[9]KALITA J,BALAS V E,BORAH S,et al.FP-Growth:A Step to Remove the Bottleneck of FP-Tree[J].Advances in Intelligent Systems and Computing,2019,176:285-296.
[10]ZHU X,LIU Y.An efficient frequent pattern mining algorithm using a highly compressed prefix tree[J].Intelligent Data Ana-lysis,2019,23:153-173.
[11]GAO Q,WAN X D.Parallel FP growth algorithm based on load balancing [J].Computer Engineering,2019,498(3):38-41,46.
[12]PEI X.Application of GRNN Neural Network in Grain Yield Prediction[J].Hans Journal of Data Mining,2020,10(4):247-253.
[13]BANI-HANI D,KHASAWNEH M.A Recursive General Re-gression Neural Network(R-GRNN) Oracle for Classification Problems[J].Expert Systems with Applications,2019,135:273-286.
[14]WANG G,ZHANG C,LIU Y,et al.A global and updatableECG beat classification system based on recurrent neural networks and active learning[J].Information Science,2018,12(3)221-228.
[15]IZONIN I,KRYVINSKA N,TKACHENKO R,et al.An Ex-tended-Input GRNN and its Application[J].Procedia Computer Science,2019,160:578-583.
[16]SONG C,WANG L,HOU J,et al.The optimized GRNN based on the FDS-FOA under the hesitant fuzzy environment and its application in air quality index prediction[J].Applied Intelligence,2021(3):1-12.
[17]FANNAS L Y,SASI A B.Off-Line Signature Recognition Based on Angle Features and GRNN Neural Networks[J].International Journal of Parallel Programming,2019,7:229-233.
[18]YANG S X,CAO Y,LIU D,et al.RS-SVM forecasting model and power supply and demand forecasting [J].Journal of Central South University,2011,18(6):2074-2079.
[19]CARNEIRO H,PEDREIRA C E,FRAN F,et al.A universal multilingual weightless neural network tagger via quantitative linguistics[J].Neural Networks,2017(11):85-101.
[1] 马力文, 周颖.
改善STARTUP阶段空窗现象的BBR单边适应算法
BBR Unilateral Adaptation Algorithm for Improving Empty Window Phenomenon in STARTUP Phase
计算机科学, 2022, 49(2): 321-328. https://doi.org/10.11896/jsjkx.201200266
[2] 朱岸青, 李帅, 唐晓东.
Spark平台中的并行化FP_growth关联规则挖掘方法
Parallel FP_growth Association Rules Mining Method on Spark Platform
计算机科学, 2020, 47(12): 139-143. https://doi.org/10.11896/jsjkx.191000110
[3] 叶俊,张正军.
基于DS-Adaboost算法的人脸检测
Face Detection Based on DS-Adaboost Algorithm
计算机科学, 2013, 40(Z11): 318-319.
[4] 余嘉元 汪存友.
基于神经网络的项目参数估计方法

计算机科学, 2008, 35(3): 134-136.
[5] 秦亮曦 李谦 史忠植.
基于排序FP-树的频繁模式高效挖掘算法

计算机科学, 2005, 32(4): 31-33.
[6] 袁鼎荣 张师超.
基于频繁链表的频繁集的挖掘算法

计算机科学, 2003, 30(7): 165-166.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!