Research And Implementation Of Key Algorithms For Big Data Oriented Education

The development trend of modern education is based on a large number of data on the basis of personalized, standardized. Data are compared with the traditional data,unstructured, distributed, the huge amount of data, data analysis by experts layer changes for user layer, use visualization to show the method of, these characteristics precisely adapted to the personalized and humanized learning changes. In the wave of the education revolution, triggered by the online education education by digital support to data to support change(environmental education, experiment scene, the temporal and spatial change and learning changes, changes in educational management and so on) is a potential goldmine. The design of environmental education, education experiment scene layout, educational space and time change, learning scene change, education and management of data acquisition and decision making these past rely on experience or ideas of things, in the cloud, networking, data under the background of, become a data support the behavioral sciences. There are two important research directions in the application of big data online education, namely, the large data mining and learning analysis technology, the main contents are as follows:1. Research learning analysis technology, were introduced and in-depth analysis of the learning of technical field leader of Siemens proposed learning model analysis, and comparison and analysis of various models, for the later chapters also lay a solid foundation.2. Research and analysis of large data mining technology, combined with the actual situation of online education in a large number of data mining to obtain useful information, to provide services for teaching optimization and educational decision-making.3. To achieve the operation in the Hadoop parallel Apriori algorithm. Through the research on segmentation of data set thinking, from the key to streamline the optimization, improve efficiency of parallel computing. Analysis of the characteristics of educational data, in view of the user-course access frequency data to do the data set processing, so as to improve the accuracy of association rules mining.4. The traditional clustering collaborative filtering algorithm is improved based on,for each knowledge point to construct the weak link of each student establishing knowledge map, according to the mapping relationship between knowledge point map analysis and knowledge and questions to students recommended for more specific questions and teaching resources.5. For online education in the automatic test paper, research and analysis of the traditional test system used in the core algorithm, based on genetic algorithm of automatic test paper system.