论文部分内容阅读
针对传统谱库检索效率低,提出了一种基于Tversky特征相似度的谱库检索算法。它优化了基于向量空间模型的谱库检索算法,并在准确定性分析化合物的前提下,通过大幅度减少原有相似度算法中的乘积运算来缩短谱库检索的时间。实验表明,该算法的检测结果和NIST 11的结果完全一样,并且谱库检索的时间缩短了5.22%,具有一定的检索优势。
Aiming at the low efficiency of traditional spectral library retrieval, a spectral library retrieval algorithm based on Tversky characteristic similarity is proposed. It optimizes the library search algorithm based on vector space model and shortens the time of library search by significantly reducing the product operation in the original similarity algorithm under the premise of accurate qualitative analysis of compounds. Experiments show that the proposed algorithm has exactly the same result as NIST 11, and the searching time of the library has been shortened by 5.22%, which has a certain retrieval advantage.