Several critical leading techniques are discuss for Deep Web database classification, TF-IDF and DOM-tree-based feature extraction model is established, improved weighted K-NN classification algorithm is proposed. This paper utilizes UIUC data-sets and WEAK platform to carry out experiments. Results of the method (precision, recall rate and F-measure measure) and that of other literatures are compared, each indicator displays good performance.%讨论若干Deep Web数据库分类准确性的前沿技术,建立基于词频和DOM树的文本特征提取方法模型,提出计算Deep Web数据库的基于权值的K-NN(K Nearest Neighbors)分类优化算法.利用UIUC提供的TEL-8数据集和WEKA平台的各类算法进行实验,并对分类精度、召回率和综合F-measure等测度上的分类结果进行比较.实验结果表明,该方法模型在3个指标上表现都较为突出.
展开▼
机译：DeepലഒരലംംംDeep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep Deep
