Research And Application Of Education Resources Semantic Analysis From Aspect Of Big Data

With the continuous development of information and communication technology,the Internet combines all kinds of professions including system industry and development of education ideology shows new features.A variety of teaching information emerged in the era of big data,traditional educational information survey and evaluation system always need a large number of artificial marking which is difficult to guarantee the efficiency and objectivity of the investigation.How to effectively extract the characteristics of network information and solve the adaptability of network teaching and accurately grasp the needs of teaching objectives is the urgent problems of educational technology in the era of big data.This topic is from the Ministry of science and technology project "content banking evaluation system",we use text analysis module to study the key words extraction algorithm and evaluation of affective tendency and applied it to the specific text resources based on big data evaluation system.The purpose is to try to change the traditional means of investigation and evaluation,solve the problems existing in the education reform and train and search in the field of education.The primary task of text evaluation is to obtain the network data based on the big data evaluation system.In order to obtain more extensive and comprehensive internet information and the information related to the mobile internet education,web crawler technology which breaks through the restrictions of various network sites is adopted to grab the internet data and access to Wed side data and mobile terminal information data.The text analysis function module is established to analyze the more extensive and comprehensive network education resources.The key words extraction technology and text orientation analysis technology are adopted in the paper to design the sub module of text analysis function as well as quantify and standardize the non-structured data and complete the related mining and evaluation work of the text data.The effective integration and classification of unstructured text data has a certain degree of obstacles in the big data environment.In the paper clustering algorithm is used to realize the effective classification of unstructured data.Using keyword extraction technology to extract key words,can realize the feature extraction of educational text resources after text analysis,but the fragmentation of key words will result in ambiguity of the original meaning of the text and mislead the judge to make an incorrect judgment.analyzing the semantic of text from two dimensions of the most frequent sets and word relevance can effectively improve the above problems based on the association rule algorithm.The evaluation of the network education resource text will have a more clear understanding based on the association rules between the word segmentation.In view of the present situation of the low efficiency of the existing affective tendency evaluation algorithm,the random forest algorithm is proposed to train a more effective and more specific emotion dictionary training set classifier based on the field of education.Eventually,the effectiveness of the key lifting model is optimized through optimizing the number of nodes and the number of forest in random forest.