基于SA-BP算法的本体概念语义相似度综合计算

doi:10.11896/jsjkx.181202351

摘要/Abstract

摘要： 不同作战部队在指挥信息系统测试评估中建立的指标存在异构问题,导致在信息交互和测试数据共享上存在较大困难。实现指标本体概念的映射和集成,建立一个统一的全局指标本体树可以有效地解决该问题,其中本体概念相似度计算的准确性至关重要。针对现有本体概念相似度计算模型中存在的精度不高的问题,提出了基于模拟退火改进BP(Back Propagation)神经网络(Simulated Annealing Back Propagation,SA-BP)算法的相似度综合计算模型。首先,对经典的基于语义距离、信息内容和概念属性的相似度计算模型进行改进,同时提出了基于概念子节点重合度的相似度计算模型;然后,采用SA-BP算法进行相似度综合计算,避免现有方法中人为确定权重的主观性和简单线性加权的不准确性问题;最后,从某作战部队不同单位建立的各异的指挥信息系统评估指标的本体概念中提取样本数据,对相似度综合计算模型进行训练测试。实验数据表明,相比于PSO-BP计算模型和主成分分析确定权值的线性加权计算模型,基于SA-BP算法的相似度综合计算模型的计算结果与专家评价结果的Pearson相关系数分别提升了0.0695和0.1351,达到了极强相关的一致性。实验数据充分说明,模拟退火算法改进的BP神经网络在训练后可以较好地收敛,在综合计算本体概念相似度时更加准确,从而有效地解决了本体概念集成的关键问题。

关键词: BP神经网络, 本体集成, 模拟退火算法, 语义相似度计算, 子节点重合度

Abstract: There are heterogeneous problems in indicators that are established by different combat forces when evaluating and testing command information systems,which leads to great difficulties in information interaction and data sharing.In order to achieve mapping and integration of indicator’s ontology-concept,building a unified global indicator ontology tree is an effective solution.In this case,the accuracy of similarity calculation for ontology-concept becomes crucial.Aiming at the problem of low accuracy in the existing ontology-concept similarity calculation model,a comprehensive simi-larity calculation model based on BP neural network algorithm which is improved by Simulated Annealing (SA-BP),was proposed.This paper first improved the classical similarity calculation models based on semantic distance,information content and conceptual attribute.Besides,a similarity calculation model in view of concept’s sub-node coincidence was proposed in order to avoid the subjectivity of artificially determined weights and the inaccuracy of simple linear weighting in existing models.At last,a training test on the comprehensive similarity calculation model was performed,while the sample data were extracted from ontology-concept of variable evaluation indicators that come from command information systems established by different departments of combat forces.Experimental data show that compared with PSO-BP calculation model and principal-component linear weighted calculation model,the comprehensive similarity calculation model based on SA-BP algorithm achieves strong correlation,since its results and its Pearson correlation coefficient of the results evaluated by experts are increased by 0.0695 and 0.1351 respectively.The experimental results verify that,after training,SA-BP algorithm can converge better and achieve higher accurate when calculating ontology-concept similarity.Hence,key issues of integration for ontology-concept can be effectively solved.

Key words: BP neural network, Ontology integration, Semantic similarity calculation, Simulated annealing algorithm, Sub-nodecoincidence

中图分类号:

TP391

许飞翔,叶霞,李琳琳,曹军博,王馨. 基于SA-BP算法的本体概念语义相似度综合计算[J]. 计算机科学, 2020, 47(1): 199-204. https://doi.org/10.11896/jsjkx.181202351

XU Fei-xiang,YE Xia,LI Lin-lin,CAO Jun-bo,WANG Xin. Comprehensive Calculation of Semantic Similarity of Ontology Concept Based on SA-BP Algorithm[J]. Computer Science, 2020, 47(1): 199-204. https://doi.org/10.11896/jsjkx.181202351

参考文献

[1]SCHADD F C,ROOS N.Word-Sense Disambiguation for Onto- logy Mapping:Concept Disambiguation using Virtual Documents and Information Retrieval Techniques[J].Journal on Data Semantics,2015,4(3):167-186.
[2]GAO W,FARAHANI M R,ASLAM A,et al.Distance learning techniques for ontology similarity measuring and ontology mapping[J].Cluster Computing,2017,20(2):959-968.
[3]RADA R,MILI H,BICKNELL E,et al.Development and application of a metric on semantic nets[J].IEEE Transactions on Systems,Man,and Cybernetics,2002,19(1):17-30.
[4]WU Z,PALMER M.Verb Semantics and Lexical Selection
[C]∥Proceedings of 32nd Annual Meeting on Association for Computational Linguistics.LasCruces,New Mexico,1994:133-138.
[5]LEACOCK C,CHODOROW M.Combining Local Context and WordNet Similarity for Word Sense Identification[M].WordNet:An Electronic Lexical Database,1998.
[6]GOBLE A,STEVENS J R,BRASS C A,et al.Investigating semantic similarity measures across the Gene Ontology:the relationship between sequence and annotation[M].Oil Bhales of the World:Pergamon Press,2003.
[7]RESNIK P.Semantic Similarity in a Taxonomy:An Information-Based Measure and its Application to Problems of Ambiguity in Natural Language[J].Journal of Artificial Intelligence Research,2011,11(1):95-130.
[8]LIN D.An Information-Theoretic Definition of Similarity[C]∥International Conference on Machine Learning.New Brunswick,NJ,1998.
[9]TVERSKY A.Features of Similarity[J].Readings in Cognitive Science,1988,84(4):290-302.
[10]WAN S,ANGRYK R A.Measuring semantic similarity using WordNet-based Context Vectors[C]∥2007 IEEE International Conference on Systems,Man and Cybernetics.Montreal,2007:908-913.
[11]ZHANG Z P,TIAN S X,LIU H Q.A Comprehensive Method for Calculating Ontology Similarity[J].Computer Science,2008,35(12):142-145.
[12]LI Y,ZA B.An approach for measuring semantic similarity between words using multiple information sources[J].IEEE Transactions on Knowledge & Data Engineering,2003,15(4):871-882.
[13]ZHANG H Y,WEN C Y,LIU D B,et al.Improved ontology-based semantic similarity calculation[J].Computer Engineering and Design,2015,36(8):2206-2210.
[14]ZHENG Z Y,RUAN C Y,LI L,et al.Research on Adaptive Synthetic Weighting Algorithm for Ontology Semantic Similarity[J].Computer Science,2016,43(10):242-247.
[15]GAO X R,XU Y Z.Research on Improved Model for Concept Similarity Computation in Domain Ontology and Application[C]∥2017 International Conference on Robots & Intelligent System (ICRIS).Huai’an,2017:257-261.
[16]HAN X R,WANG Q S,GUO Y,et al.Semantic similarity measure of geographic ontology based on PSO-BP algorithm[J].Computer Engineering and Applications,2017,53(8):32-37.
[17]GUO X H,PENG Q,DENG H,et al.WordNet word similarity calculation based on edge weight [J].Computer Engineering and Application,2018,54(1):172-178.
[18]FAN M,ZHANG Y,LI J.Word similarity computation based on HowNet[C]∥International Conference on Fuzzy Systems & Knowledge Discovery.IEEE,2016.
[19]LI Y,GAO D Q.Research on Entity Similarity Computation in Knowledge Map [J].Chinese Journal of Information Science,2017,31(1):145-151,159.
[20]WANG S,NA Z,LEI W,et al.Wind speed forecasting based on the hybrid ensemble empirical mode decomposition and GA-BP neural network method[J].Renewable Energy,2016,94(1):629-636.
[21]METROPOLIS N,ROSENBLUTH A W,ROSENBLUTH M N,et al.Equation of State Calculations by Fast Computing Machines[J].The Journal of Chemical Physics,2004,1087(1953):21.
[22]KIRKPATRICK S,VECCHI M P.Optimization by simulated annealing[M].Spin Glass Theory and Beyond:An Introduction to the Replica Method and Its Applications,1987.
[23]MAMANO N,HAYES W B.SANA:Simulated Annealing far outperforms many other search algorithms for biological network alignment[J].Bioinformatics,2017,33(14):1-9.
[24]ZHOU A W,ZHAI Z H,LIU H T.An improved BP neural network algorithm based on simulated annealing algorithm [J].Microelectronics and Computer,2016,33(4):144-147.
[25]DE WINTER J C,GOSLING S D,POTTER J.Comparing the Pearson and Spearman Correlation Coefficients Across Distributions and Sample Sizes:A Tutorial Using Simulations and Empirical Data[J].Psychological Methods,2016,21(3):273.

相关文章 15

[1]	徐佳楠, 张天瑞, 赵伟博, 贾泽轩. 面向供应链风险评估的改进BP小波神经网络研究 Study on Improved BP Wavelet Neural Network for Supply Chain Risk Assessment 计算机科学, 2022, 49(6A): 654-660. https://doi.org/10.11896/jsjkx.210800049
[2]	朱旭辉, 沈国娇, 夏平凡, 倪志伟. 基于螺旋进化萤火虫算法和BP神经网络的模型及其在PPP融资风险预测中的应用 Model Based on Spirally Evolution Glowworm Swarm Optimization and Back Propagation Neural Network and Its Application in PPP Financing Risk Prediction 计算机科学, 2022, 49(6A): 667-674. https://doi.org/10.11896/jsjkx.210800088
[3]	刘宝宝, 杨菁菁, 陶露, 王贺应. 基于DE-LSTM模型的教育统计数据预测研究 Study on Prediction of Educational Statistical Data Based on DE-LSTM Model 计算机科学, 2022, 49(6A): 261-266. https://doi.org/10.11896/jsjkx.220300120
[4]	夏静, 马中, 戴新发, 胡哲琨. 基于BP神经网络的智能云效能模型 Efficiency Model of Intelligent Cloud Based on BP Neural Network 计算机科学, 2022, 49(2): 353-367. https://doi.org/10.11896/jsjkx.201100140
[5]	高士顺, 赵海涛, 张晓瀛, 魏急波. 一种自适应于不同场景的智能无线传播模型 Self-adaptive Intelligent Wireless Propagation Model to Different Scenarios 计算机科学, 2021, 48(7): 324-332. https://doi.org/10.11896/jsjkx.201000181
[6]	王国武, 陈元琰. 基于跳数修正和遗传模拟退火优化DV-Hop定位算法 Improvement of DV-Hop Location Algorithm Based on Hop Correction and Genetic Simulated Annealing Algorithm 计算机科学, 2021, 48(6A): 313-316. https://doi.org/10.11896/jsjkx.201000101
[7]	郭福民, 张华, 胡瑢华, 宋岩. 一种基于表面肌电信号的腕部肌力估计方法研究 Study on Method for Estimating Wrist Muscle Force Based on Surface EMG Signals 计算机科学, 2021, 48(6A): 317-320. https://doi.org/10.11896/jsjkx.200600021
[8]	程铁军, 王曼. 基于变权组合的突发事件网络舆情趋势预测 Network Public Opinion Trend Prediction of Emergencies Based on Variable Weight Combination 计算机科学, 2021, 48(6A): 190-195. https://doi.org/10.11896/jsjkx.200600094
[9]	石琳姗, 马创, 杨云, 靳敏. 基于SSC-BP神经网络的异常检测算法 Anomaly Detection Algorithm Based on SSC-BP Neural Network 计算机科学, 2021, 48(12): 357-363. https://doi.org/10.11896/jsjkx.201000086
[10]	周俊, 尹悦, 夏斌. 基于LSTM神经网络的声发射信号识别研究 Acoustic Emission Signal Recognition Based on Long Short Time Memory Neural Network 计算机科学, 2021, 48(11A): 319-326. https://doi.org/10.11896/jsjkx.210700034
[11]	焦东来, 王浩翔, 吕海洋, 徐轲. 基于手机传感器轨迹的路面地物检测方法 Road Surface Object Detection from Mobile Phone Based Sensor Trajectories 计算机科学, 2021, 48(11A): 283-289. https://doi.org/10.11896/jsjkx.210200145
[12]	宋岩, 胡瑢华, 郭福民, 袁新亮, 熊睿洋. 基于sEMG的改进SVM+BP肌力预测分层算法 Improved SVM+BP Algorithm for Muscle Force Prediction Based on sEMG 计算机科学, 2020, 47(6A): 75-78. https://doi.org/10.11896/JsJkx.190900143
[13]	诸珺文. 基于改进BP神经网络的SQL注入识别 SQL InJection Recognition Based on Improved BP Neural Network 计算机科学, 2020, 47(6A): 352-359. https://doi.org/10.11896/JsJkx.191200054
[14]	周立鹏, 孟利民, 周磊, 蒋维, 董建平. 基于BP神经网络的摔倒检测算法 Fall Detection Algorithm Based on BP Neural Network 计算机科学, 2020, 47(6A): 242-246. https://doi.org/10.11896/JsJkx.191000077
[15]	陈燕文,李坤,韩焱,王燕平. 基于MFCC和常数Q变换的乐器音符识别 Musical Note Recognition of Musical Instruments Based on MFCC and Constant Q Transform 计算机科学, 2020, 47(3): 149-155. https://doi.org/10.11896/jsjkx.190100224

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed