Recognition And Discovery Of Programing Design Network Resource Named Knowledge Entity

With the rapid development of Internet technology,mankind has been accustomed to acquire knowledge from the network.But with the explosive growth of network resources and the diversity of network resources,the method of people to obtain knowledge from the browser is stagnant.So,a tool is needed to help people from the network to obtain and discover new knowledge efficiently.Because the text of network resource is not completely structured data,it meanwhile includes complex unstructured data,such as free text.Although this text information is convenient for people to express concepts and events freely,it also makes obstacle for machine search and statistical analysis.Named Entity Recognition(NER),as one of the basic tasks in natural language processing,plays a fundamental role in automatic question-answering and information extraction.This paper regards NER as a technical entry to study these issues.The paper explores and studies in the following aspects:First of all,on the basis of investigating the common naming entity recognition solutions at home and abroad,this paper analyzes the development trend and technical characteristics of the mainstream model at the present stage.This paper summarizes the limitations of current mainstream machine learning methods based on statistics,and considers the characteristics of neural network on textual deep semantic and semantic mining,and points out the new idea of building neural network to study the naming knowledge entity recognition.Secondly,a window-based deep neural network model is proposed,and the application of the model in naming entity recognition task is studied deeply.This paper solves the problem of transformation from network resources to model input,and deduces forward-backward propagation algorithm.At the same time,we give a detailed introduction to the parameters in the model frame and give the parameters initialization and tuning techniques to enhance the labeling effect.Thirdly,in the field of algorithmic knowledge,a large number of research and experiments on the extraction and discovery of knowledge entities have been carried out.The experimental results show that the neural network model has good recognition effect of the algorithm knowledge,the accuracy of the recognition algorithm is 98%,and the new knowledge points outside the expert database can be found effectively to realize the expected experimental requirements.At last,the system of knowledge entity recognition based on deep neural network is realized,and the non-linear processing of the model is improved.At the same time,the model algorithm is optimized,which solves the problem of too long training and high maintenance cost in practice,which makes it more suitable for dealing with massive data.It is proved that the neural network model based on depth learning can be applied to the field of algorithm knowledge well.The algorithm is based on the model as the core.The entity recognition discovery system has good maintainability and robustness,and meets the new requirements of algorithm knowledge entity recognition in large data background.