Automatic Search Method for Unstructured Database of Natural Language Instructions Combining Cross-Step Matching Word Segmentation Algorithm and Web Text Clustering Analysis Model

In order to search for effective information from massive and complex databases, an automatic search method for unstructured databases of natural language instructions is proposed, which combines cross-step matching word segmentation algorithm and Web text clustering analysis model. Firstly, natural language processing operations such as word segmentation and part-of-speech tagging are carried out on the input natural language instructions through the cross-step matching word segmentation algorithm, and the natural language instructions are transformed into vectors that can be processed by the computer through the support vector machine deep denoising self-encoder. Web text clustering algorithm is used to cluster the unstructured database, and similar text data are normalized to the same cluster matrix to extract the unstructured data features of natural language instructions. Through keyword extraction and semantic analysis of clustering results, the clustering results are trained by machine learning technology to realize automatic search of database information. Experimental results show that the search performance of the proposed method is better, the search similarity is close to 99%, and it has a high search effect.