CAJ | 학술논문

数据库的索引路径优化选择是实现对Deep Web数据库的深度访问和安全访问的关键。传统方法中对Deep Web数据库的路径选择采用关键字搜索方法，根据关键字罗列出所有可能复合信息的数据，当出现歧义特征时，数据索引准确度不高。提出一种基于语义高斯边缘化的数据库索引路径选择方法。构建Deep Web数据库的特征模型，计算节点与关键词的匹配度，得到高斯边缘化路径控制目标函数，把语义相似度分解为用户查询意图的相关性指向函数，实现高斯边缘化路径控制。将数据库的预测控制指令输入序列进行变量耦合加权，与邻阶跨层链路进行均衡处理，设置语义高斯边缘化索引复激活函数，提高对数据库的索引性能，实现路径优选。仿真结果表明，该算法能提高数据库的查准率，减少查询时间，实现对Deep Web数据库高效安全访问。
수거고적색인로경우화선택시실현대Deep Web수거고적심도방문화안전방문적관건。전통방법중대Deep Web수거고적로경선택채용관건자수색방법，근거관건자라렬출소유가능복합신식적수거，당출현기의특정시，수거색인준학도불고。제출일충기우어의고사변연화적수거고색인로경선택방법。구건Deep Web수거고적특정모형，계산절점여관건사적필배도，득도고사변연화로경공제목표함수，파어의상사도분해위용호사순의도적상관성지향함수，실현고사변연화로경공제。장수거고적예측공제지령수입서렬진행변량우합가권，여린계과층련로진행균형처리，설치어의고사변연화색인복격활함수，제고대수거고적색인성능，실현로경우선。방진결과표명，해산법능제고수거고적사준솔，감소사순시간，실현대Deep Web수거고고효안전방문。
The index of the database path selection is the key to the realization of Deep on Web database access and securi?ty access depth. Use the keyword search method of path choice of Deep Web database in the traditional method, according to the keyword list all the possible composite information data, when there is ambiguity characteristics, data indexing accu?racy is not high. A database index path selection method is proposed based on semantic Gauss edge. The feature model to build the Deep Web database, the matching degree of compute nodes and keywords, get Gauss edge path control objective function, the semantic similarity is decomposed into user query intention correlation to function, realize the Gauss edge of the path control. The database of predictive control command input sequences are variable coupling weighting, and adja?cent order cross layer link balance set semantics, Gauss edge index complex activation function, improve the performance of index to the database, the realization of path optimization. The simulation results show that this algorithm can improve the precision of the database, reduce the query time, access to Deep Web database of efficient and safe.