CAJ | 학술논문

为了更好地组合不同的相似度度量结果以提高本体映射结果的质量，提出一种新的基于调谐值度量和单纯降序提取算法的自动化本体映射技术。该技术首先通过调谐值来度量各种相似度矩阵的可靠性，并为每一个相似度矩阵赋予权重以集成不同的相似度矩阵，然后通过单纯降序提取算法结合阈值的策略提取最终的本体映射结果。实验采用2012年本体映射评价竞赛的测试数据集，同2012年本体映射评价竞赛的参与者的比较表明该文提出的方法是有效的。
위료경호지조합불동적상사도도량결과이제고본체영사결과적질량，제출일충신적기우조해치도량화단순강서제취산법적자동화본체영사기술。해기술수선통과조해치래도량각충상사도구진적가고성，병위매일개상사도구진부여권중이집성불동적상사도구진，연후통과단순강서제취산법결합역치적책략제취최종적본체영사결과。실험채용2012년본체영사평개경새적측시수거집，동2012년본체영사평개경새적삼여자적비교표명해문제출적방법시유효적。
To improve the quality of ontology alignment through better aggregating the alignments ob-tained by various similarity measures,a novel automatic ontology aligning technology based on the tuning measure and naive descending extraction algorithm was proposed.Firstly,the tuning value was utilized to measure the reliability of each similarity matrix,and each similarity matrix was given a weight to aggregate them.Then both naive descending extraction algorithm and threshold strategy were used to extract the final ontology alignment.The testing cases from the ontology alignment eval-uation initiative in 2012 were adopted as the data group.The comparison with the participants’algo-rithm of ontology alignment evaluation initiative in 2012 shows the effectiveness of the proposal.