About the author: Yang Xianmin (1982- ), male, Xingtai, Hebei, associate professor of Jiangsu Education Information Technology Engineering Research Center, Jiangsu Normal University, mainly engaged in wisdom education, mobile and ubiquitous learning research, E-mail:; Wang Liuhui, Jiangsu Normal University Jiangsu Education Information Engineering Technology Research Center, Xuzhou, Jiangsu 221116, China; Tang Sisi, National Information Center, Beijing 100045, China
Summary: Big data is the scientific force that promotes the development of educational innovation. Education Big Data is a collection of data that is generated throughout the educational activity and that is collected according to educational needs for educational development and that creates enormous potential value. Compared with traditional education data, the collection of educational big data has stronger real-time, coherence, comprehensiveness and naturalness. The analysis and processing are more complicated and diverse, and the application is more diverse and deeper. The five-tiered structure of educational big data includes individual-level data, course-level data, school-level data, regional-level data, and country-level data, and aggregates various educational data from the bottom up. Since the 1970s, the application of educational data has generally experienced the initial start-up phase, the key exploration phase, and the rapid development phase. The scale and level of educational data application in the three phases have continued to evolve. The application of educational big data is mainly embodied in: driving the national education policy scientifically; driving the balanced development of regional education; driving the improvement of school education quality; driving the optimization of curriculum system and teaching effect; driving the individualized development of individuals. Finally, in view of the problems and challenges in the current development of China's educational big data, six policy recommendations were put forward: the introduction of "Educational Big Data Application Development Guidance Opinions", the "Education Big Data Security Management Measures", and the establishment of the National Education Big Data Research Institute. Established the National Education Big Data Governance Body, promulgated the “Educational Data Operator†license, and accelerated the construction of the Education Big Data Industry Base.
Keywords: education big data education data education development application model policy recommendations
Title Note: Jiangsu University's Advantageous Subject Construction Project Funded Project “Jiangsu Normal University Education Department Advantage Discipline Construction†(Su Zheng Ban Fa [2014] No. 37); National College Student Practice Innovation Training Program Project “Big Data Supported Learning Behavior Record Research on the use of developmental evaluation (item number Z).
The era of technological change education has arrived, and the modernization of education driven by information technology has risen to the national strategy. [1][2] At present, China's education development faces many problems (burden reduction, fairness, quality improvement, balanced development, etc.), while the rapid development of information technology such as cloud computing, big data, learning analysis, Internet of Things, mobile communication, etc. Educational difficulties and the promotion of comprehensive reform and development in the field of education provide important opportunities and great possibilities. Among them, big data technology is undoubtedly the scientific force to promote the development of educational innovation. In recent years, big data has continuously exerted profound influence on various fields of society, and is undergoing major changes in human work, life and thinking. Similarly, its "power" has also strongly impacted the entire education system and is becoming a subversive force driving innovation and change in the education system.
Education big data is a subset of big data, especially big data in the field of education. It is a collection of data generated during the entire educational activity and collected according to educational needs, all used for educational development and creating great potential value. . The “big†of education big data is not only the quantity, but the emphasis on “valueâ€, that is, it can find relevant relationships from the complicated educational data, diagnose existing problems, predict development trends, and promote the education big data. The important role of education quality, promoting education equity, realizing individualized learning, optimizing the allocation of educational resources, and assisting scientific decision-making in education. At present, big data has attracted the attention of educational researchers, managers, policy makers and practitioners. Some scholars have explored the impact of big data on the development of education from the transformation of educational model, [3] the possible turn of education, [4] the new paradigm of educational research, [5] the change of learning methods [6]. Some scholars have explored the methods and applications of learning analysis and educational data mining from a technical perspective. [7] [8] However, there is still a lack of systematic review and analysis of the data categories and application models of educational big data.
Based on this, this paper aims to explore four aspects: What are the characteristics of educational big data, and what types of data are involved? What are the stages in the application process of educational data? The application value and mode of educational big data in education is What? What problems are faced in the development of education big data, how to solve it?
First, the characteristics and structure of educational big data
Compared with traditional education data, the collection of educational big data has stronger real-time, coherence, comprehensiveness and naturalness. The analysis and processing are more complicated and diverse, and the application is more diverse and deeper. The collection of traditional educational data is often staged, mostly in the case of the user's knowledge (unnatural state). The means of analysis mostly use simple summary statistics and comparative analysis. The focus of attention is on the group characteristics of the educated and the country. The overall situation of education development at different levels, regions and schools. In the era of big data, new technologies such as mobile communication, cloud computing, sensors, and pervasive computing will gradually be integrated into the whole process of education, and more microscopic teaching and learning can be collected in real time and continuously without affecting the teaching activities of teachers and students. Process data, such as the student's learning trajectory, the time spent on each assignment, the number of questions and smiles in the teacher's classroom. The data structure of educational big data is more mixed. Regular structured data (such as grades, student status, employment rate, attendance records, etc.) is still important, but unstructured data (such as pictures, videos, lesson plans, teaching software, learning games, etc.) Will become more and more dominant.
Figure 1 "Iceberg Model" of Education Big Data
Educational data is generated all the time, but what data is included in the education field? What data needs to be collected? Drawing on the “iceberg model†of human resources in the field of human resources, we can build an “iceberg model†for educational big data (as shown in Figure 1). Show). The model divides the education data into two parts, the data that is exposed above the ice and the data that is hidden under the ice. Over the years, the educational data collected by the state is mainly based on management, structure and results. These data are located above the “ice surface†and are characterized by easy measurement and explicitness, focusing on the overall situation of education development at the macro level. In a certain historical period, it played a positive role in formulating education policies and promoting education development in China. However, with the advent of the era of big data, the international community has increasingly recognized and valued the status of educational big data strategic assets, and the comprehensive collection and deep mining analysis of educational data has become more and more important. The focus of educational data collection will shift to unstructured, procedural data, which is mainly located below the “ice surface†and is characterized by difficulty in measurement and recessiveness. These data will far exceed traditional educational data in terms of quantity, growth rate, and potential value.
Educational data is objective, and its value depends on the people who manipulate and apply the data. Therefore, both the data above the ice and the data below the ice are important components of educational big data. Only from the current collection and application of educational data, we should focus on strengthening the collection and deep mining of some educational data under the ice, and strengthen the integration of educational big data and other fields of big data (medical, transportation, economic, social security, etc.). And correlation analysis to further enhance the scientific nature of educational decision-making. In order to more clearly understand the overview of educational big data, here according to the source and scope of educational data, it is divided into five layers of structure (as shown in Figure 2), from the bottom up to a variety of educational data.
Figure 2 Hierarchical architecture of educational big data
Individual level education data: including the basic information collected by the state and the students, and various behavior data of the user (such as the student's learning behavior record anytime and anywhere, the management's various operational behavior records, the teacher's teaching behavior record, etc.) User status description data (such as learning interests, motivation, health status, etc.).
Course level education data: relevant educational data generated around the course teaching, including basic information of the course, course members, curriculum resources, coursework, teacher-student interaction, course assessment, etc., where the course member data comes from the individual layer and is used for Describe personal information related to student course learning.
School level education data: mainly includes various school management data (general profile, student management, office management, scientific research management, financial management, etc.), classroom teaching data, educational data, campus safety data, equipment use and maintenance data specified by national standards, Use data such as classroom laboratories, school energy consumption data, and campus life data.
Figure 3 Education big data panorama
Regional level education data: mainly from schools and social training and online education institutions, mainly including education administrative management data stipulated by national standards, various behavior and result data generated by regional education cloud platform, and regional teaching and research training. Educational resources, data on teaching and research and student competition activities at various regional levels, and various social training and online educational activities.
National level education data: mainly aggregates various educational data generated from various regions. Figure 3 shows the breakdown data types included in each layer of education data.
Second, the development stage of educational data application
With the development of artificial intelligence, data mining, machine learning and other technologies, the data processing and application in the education field has gradually matured. Since the 1970s, the development stage of educational data application has generally experienced the initial start-up phase, the key exploration phase and the rapid development phase. The scale and level of educational data application in these three phases are continuously progressive.
(1) Initial start-up phase (1970-1997)
Artificial intelligence was born in the mid-1950s. After more than a decade of development, it began to enter the field of education in the 1970s. With the support of AI technology, the traditional CAI went to ICAI and subsequently developed into the intelligent tutor system ITS. ITS mainly includes expert models, student models, instructional models and guidance environment subsystems, [9] through the collection and analysis of learners' learning methods, learning habits and learning processes to improve learning methods and improve learning efficiency. It should be said that at the time, researchers had high expectations for the application level of educational data, and there were clear application ideas. Unfortunately, due to the high cost of computer use and poor performance at the time, it has greatly hindered the application of this method in the field of education.
In the mid-1980s, data warehousing technology began to emerge. Due to its topic-oriented, integrated, time-varying and non-volatile characteristics, it has gradually become an important platform for data analysis and online analysis. [10] Data warehouse provides effective support for a wide range of strategic decisions and long-term trend analysis for various educational institutions by providing historical data for different types of application systems. In the late 1980s and early 1990s, data mining and knowledge discovery became an active research field. Various professional international conferences were held one after another, which set off a research and application boom in data mining and knowledge discovery in various fields. Researchers in the field of education have also begun to pay attention to data mining and knowledge discovery technology, and have made preliminary attempts and explorations. However, due to the limited development level of networks and computers, there has been no substantial progress in the exploitation and utilization of educational data.
In general, computers at this stage were initially introduced into the field of education, and the application range and degree of various advanced technologies were relatively low, and the educational data collected was relatively small. The biggest progress in this phase is the realization of the digitization of educational data, from the simulation to the digital provides a good condition for the permanent storage and rapid dissemination of educational data. Secondly, the improvement of educational data on teaching has begun to attract the attention of the academic community. However, due to the lack of computer popularity and the lag of database technology, the application value of educational data in the field of education has not been reflected, and the application level is at an early stage.
(II) Key exploration stage (1998-2007)
In 1998, "Science" magazine published an article on the computer software HiQ "Big Data Processing Program", the first use of the word big data. However, because the development capability of the IT industry and the industrial use of information resources were still in their infancy at the time, the concept of big data was in its infancy and did not receive the attention it deserved.
In 2002, the United States passed the Education Science Reform Act, which clarified the decisive position of data in education decision-making. "All educational policies must be formulated with empirical data." After years of silence, artificial intelligence technology was launched in the field of higher education in 2004. The “intelligent tutor system†and “artificial intelligence system†once again became popular, and a research boom of “educational data mining†was launched, which also promoted learning. The birth of analytical technology. [11] Since 2005, several “Educational Data Mining†seminars have been held at international conferences such as artificial intelligence, artificial intelligence education applications and intelligent tutor systems. The International Education Data Mining Working Group and the International Education Data Mining Association have also Established one after another. Educational data mining mainly studies the application of data mining technology in the field of education. From the perspective of its research field, educational data mining research includes “application in teaching research†and “application in educational administrationâ€. [12]
At this stage, computer and network technologies are developing rapidly, and they are widely used in the field of education. Countries around the world have stepped up efforts to build and promote educational informationization. [13][14] Various educational information application systems (such as learning management systems, educational management systems, personnel management systems, teaching resources, etc.) have emerged. The digitization and networking of various educational businesses promote the generation and storage of massive educational data (learner information, learning process data, learning result data, teaching management data, etc.). The biggest advancement in the application of educational data at this stage is that educational data mining has begun to take a lot of attention from multidisciplinary researchers, such as computers, information management, and education, as a specialized research field, and has developed rapidly. In addition, the quantity and efficiency of educational data collection has also increased significantly. The shortcoming is that due to the lack of understanding of the strategic value of educational data and the fact that educational data mining technology is still in the development stage, the application of educational data in the field of education is still not fully reflected. Most educational data is still used for simple statistics. Analysis to assist education and teaching management.
(3) Rapid development stage (2008-present)
In September 2008, the "Big Data" special issue of "Nature" magazine first proposed the word big data. After more than three years of "fermentation", big data caused unprecedented attention in 2012. The amount of global data has entered the ZB era, and various types of data have grown exponentially, and the era of big data has arrived. Big data technology continues to have a profound impact on all areas of society, and is undergoing major changes in human work, life and thinking. [15] In the context of the era of big data, countries around the world have stepped up their efforts in the field of big data in education, and introduced relevant policies and documents. Data-driven education reform and development is the general trend. In order to promote the application of “Big Data†education, the US Department of Education released “Promoting Teaching and Learning through Educational Data Mining and Learning Analysis†in October 2012, pointing out that it is important to develop educational data mining and learning analysis techniques through the education of big data. The excavation and analysis promotes the transformation of the teaching system of American colleges and universities and K-12 schools [16]. At present, the Ministry of Education of China is vigorously promoting the construction of two-level (national and provincial) educational data centers, and achieving unified and standardized management of national education data through “two-level construction and five-level applicationâ€. Some cities and districts in China are also relying on the construction and development of smart cities and smart education to vigorously develop the construction and innovation of regional education big data center platforms.
The arrival of the era of big data has pushed the development of educational data to the "fast lane." Cloud computing, Internet of Things, big data, mobile communications and other new information technologies have begun to take root in the fertile soil of education. Educational data has “explosive†growth, and the collection of educational data is more real-time, coherent, comprehensive and natural. The analysis and processing of educational data is more complicated and diverse, and the application is more diversified and in-depth. In addition, the value of educational data as an important asset has been gradually recognized and valued. Education data mining and learning analysis technology has been greatly developed and is being widely applied to various business fields such as teaching, management, research, evaluation, and service. The scale effect is Prominent. The application of educational data has entered a new historical period, based on educational data mining and learning analysis technology, research and development of specialized educational data analysis decision models, tools and algorithms to achieve the high efficiency of educational data processing and the maximum value of data applications. The outline of the data analysis and application system of the education industry is gradually clear.
Third, the application mode of education big data
Education big data is an intangible asset. It is a “gold mine†that can be mined indefinitely. Full mining and application is the only way to realize the value-added of data “assetsâ€. In a sense, the future international education competition will be the confrontation on the stage of educational big data. Then, how does education big data work? What is the application model? Next, we will discuss the application model of big data in education from different levels based on the hierarchical structure of education big data proposed above.
(1) Education big data drives the nationalization of national education policy
Data plays a key role in the formulation of national education decisions, and the arrival of the era of big data will make data collection and analysis more convenient, fast, comprehensive and accurate, and the formulation of educational policies will rely more on data.
The collection channels and quantity of traditional education data are very limited. Only through simple statistical analysis can reflect the partial situation of national education development in a certain period of time. The predicted value of data is difficult to play and cannot provide scientific support for the formulation of national education policy. Big data technology has the advantages of data quantification, path diversification, and deepening of mining. It can discover the intrinsic relationship between various educational data and other social industry data, and help to build a more systematic educational development model. Promote the scientificization of national education policy formulation and adjustment. In addition, data-based education decisions can enhance the understanding and support of education policies. [17]
The United States was the first country to determine the strategic position of educational data and to enact the relevant safeguards bill (the Education Science Reform Act). As early as 2002, the US government clearly stated in the form of legislation that all educational reforms and decisions must be supported by empirical data. . The scientific decision-making will make the overall cost of education show a downward trend, while at the same time achieving a significant increase in the quality of education and education equity. At present, the “Lifetime One Person One†electronic student registration management method that the students are pursuing in China provides institutional guarantee for the continuous recording of each student's academic performance and overall development. If we can establish a national network of student growth archives, supplemented by diverse data on families, teachers, and schools, this will be the current national college entrance examination admission and admission policies, student employment policies, resource allocation policies, and student choice policies. The improvement and optimization of education policy provides the most valuable data support.
(II) Education big data drives the balanced development of regional education
The balanced development of regional education is a major practical issue facing China's education. The application of big data technology can accurately grasp the developmental dynamics of regional education and the key factors affecting its balanced development, and comprehensively promote the balanced development of regional education from the aspects of balanced educational environment, balanced educational resources, equal educational opportunities, and balanced education quality. [18] In addition, different regions have different educational status quo. In the context of big data, not only can the educational gap between regions be narrowed, but also different regions can form different regions according to their own environmental conditions, economic conditions and development needs. Educational development path.
In fact, some international organizations have begun to apply big data technology to the allocation of educational resources. The latest EFA Global Monitoring Report released by UNESCO shows that valuable funds for education around the world are being wasted by low-quality education, with losses of up to $129 billion per year. In this regard, the report calls on governments to focus on the quantity and quality of teachers when investing in education, and to ensure that the best teachers are assigned to the students who need them the most. [19] Big data can help education sector leaders and decision makers understand the impact of funding policies. Tod R. Massa, head of policy research and data warehousing at the Virginia State Higher Education Commission, said: "Our goal is to create an environment — - All discussions are based on facts, not on intuition or conjecture." [20]
The unified student information management system provides an important guarantee for the collection, management and application of educational big data in China. Students' education, transfer, suspension, and withdrawal of education management data can achieve comprehensive, real-time collection, monitoring, updating and analysis. Education management data can also be correlated with household income, household registration, medical care, insurance, transportation and other data to help early detection and prediction of students with learning difficulties, school choices, etc., who need educational assistance and intervention, and then provide targeted Education support services to ensure that every student has equal access to quality education.
In addition, by establishing a continuous institutionalized regional education development data collection mechanism, it is possible to comprehensively track the learning situation of all students and the work situation after graduation, and then more objectively evaluate the quality of regional education, and dynamically adjust regional education according to the evaluation results. The system, such as professional adjustment, curriculum adjustment, and adjustment of training methods, achieve a more seamless connection between education and social needs, helping each student to succeed.
(3) Education big data drives the quality of school education
With the emergence of various wisdom teaching and management platforms, the role of big data in the transformation of school education will continue to be prominent. Big data has unique advantages in improving the quality of school management and teaching quality as well as improving educational evaluation methods.
The construction of digital campus has greatly promoted the digitalization and networking of school management. [21] Office automation system, asset management system, educational management system, scientific research management system and other application systems provide real-time collection and deep mining of education management data. condition. At present, some universities in China have taken the lead in carrying out education management services based on big data. Zhejiang University systematically collects and organizes the school's equipment asset data, provides convenient inquiry and analysis services, and improves the utilization and management efficiency of laboratories, classrooms, instruments, equipment and other resources. Jiangnan University has comprehensively monitored and optimized the data of school water and electricity through the Internet of Things technology to achieve energy conservation and environmental protection. East China Normal University uses students' food and beverage consumption data to provide emotional comfort and bursary support to students with financial difficulties. In addition, big data can also play an important role in teacher recruitment. By analysing and forecasting the personal information of the candidate, it is possible to recruit teachers who are more likely to succeed and become more suitable. Some school districts in the United States have begun working with big data companies to apply big data tools to assist teachers in recruiting. Through the analysis of the teacher's degree and expertise as well as the beliefs, outlook on life, attitude, openness of experience and other factors, combined with the interview results, comprehensively determine whether the teacher is hired.
In addition to the improvement of the quality of school management services, the analysis and prediction of massive teaching data through the application of big data technology will also change the traditional one-size-fits-all teaching mode to achieve high-quality, personalized teaching. Big data can fully record the learner's growth record and conduct scientific analysis, so that learners can better understand themselves, help teachers predict student achievement, and provide learners with scientific learning suggestions to help them improve their academic performance. Canada's Desire2Learn Technology has developed a “Student Success System†for colleges and universities. The system is based on the student's existing academic performance data to predict and improve its performance in future course learning, and present the results to the teacher in detail. Teachers provide personalized guidance.
Big data technology also provides comprehensive data support for students' academic achievement evaluation. All of the student's learning process and results data will be stored in the learning portfolio, and all of the teacher's teaching process and results data will be stored in the teaching portfolio. Based on portfolio data, a scientific development model can be established to regularly assess the development of students and teachers and provide corresponding development recommendations. In addition, big data technology is also useful in school scientific research activities. On the one hand, educational data can be used as a data source for educational research. Through deep data mining, it is possible to discover the essence through the appearance of educational problems and produce high-quality empirical research results. On the other hand, the analysis of massive and multi-dimensional data can play a role. The role of the Scientific Research Compass helps researchers to accurately grasp the frontier issues and trends in the research field.
(4) Optimization of educational big data-driven curriculum system and teaching effects
At present, the state is actively promoting the transformation of some undergraduate colleges to applied universities in order to meet the challenges of social development for higher education. A core issue in the transformation process is to open a major and how to set up a curriculum. At this time, it is necessary to effectively integrate the professional, industry, regional economic and social development and other data, through comprehensive data collection and in-depth analysis, accurately grasp the market demand for talents, clarify the objectives of each application-oriented professional training, build capacity Quality model, matching the appropriate curriculum system.
In fact, education departments and schools of all levels in China have accumulated a large number of student enrollment, graduation and curriculum setting data through the student status management system, educational management system, and degree management system, but these data are basically in a “sleep†state. Relevant analysis of the school's previous grades, teaching methods, employment status and other data can identify key factors affecting the course performance and employment, and provide a reliable basis for the adjustment of the school curriculum and the optimization of the teaching methods. [twenty two]
Through big data technology, it is also possible to continuously track the teacher's teaching process, analyze the teacher's teaching characteristics and its advantages, determine which teaching tasks the teacher is suitable to undertake, and what method can be used to achieve the optimal teaching effect. Through the intelligent network teaching platform, teachers can accurately judge the students' interest points, knowledge base, learning preferences, learning difficulties, etc. in advance, so as to carry out targeted teaching. Based on the student's online learning data, a variety of predictive models can be constructed to predict possible students with learning difficulties and risk of dropping out of school, and timely intervention. Purdue University's “Course Signals†project is one of the internationally renowned examples of big data education applications. [23] The project developed a course learning early warning platform, collected a large number of student process learning data, and analyzed the probability of student course success through a set of prediction algorithms. According to the prediction results, teachers can give targeted help, guidance and feedback. Significantly improved the success rate of course learning.
In addition, by analyzing the trajectories of students' clicks, browsing, page turning, collection, evaluation, etc. on the curriculum resources, it is possible to objectively evaluate the attention of the curriculum resources, the rationality of the resource interface design, the efficiency of resource navigation, and the learning effect. The impact of the curriculum resources to improve the structure and content of the curriculum resources, to achieve the generation and evolution of a large number of quality curriculum resources.
(5) Education big data drives the individualized development of individuals
The current standardized talent training model can no longer meet the development requirements of the information society, and big data provides conditions and opportunities for the innovation of talent training model. From popularization to individualization has become an important trend in the future talent training. Students and teachers are the two core subjects of talent training in the field of education. Big data will truly promote the individualized development of students and teachers.
The premise of individualized development is that individuals must first truly understand themselves, know their own strengths, weaknesses, interests, preferences, styles, knowledge defects, ability defects, development goals, etc. Secondly, they need to provide the environment, resources and activities that are most suitable for individual development. External conditions such as tools and services. The biggest advantage of big data is that students and teachers can understand each real “selfâ€. At the same time, through the deep exploration and analysis of learning behavior and teaching behavior data, the most suitable learning resources are pushed for each real “selfâ€. Learning path.
Although e-learning has the natural "personalization" advantage, but lacks the support of big data, the machine will not be able to truly understand each learner, and it will not be able to push personalized resources and services. If the Internet promotes the democratization of education, then big data will personalize education. [24] With the support of big data, teachers can pay more attention to each student individual, record the learning trajectory of each student, analyze each student's learning behavior, predict their learning results, diagnose their learning needs and problems, and carry out Really teach students in accordance with their aptitude. Teachers are gradually transformed from educators to instructors who help each student to personalize their learning and development. The traditional learning management system will be upgraded to a smart learning platform, which can continuously collect learner's learning behavior data, perform intelligent analysis, push appropriate learning resources according to the learner model, accurately diagnose and evaluate the learning process and results, and feed back to learners. The most suitable learning advice to achieve individualized development of each student.
In the field of big data to promote the professional development of teachers, some scholars pointed out [25]: Education big data can improve the learning efficiency of teachers as online learners, stimulate their independent professional development awareness; can improve the teaching efficiency of teachers as online teachers, Develop the wisdom of online teaching practice; improve the research performance of teachers as researchers, improve their service ability to students' online learning; improve the management efficiency of teachers as managers and enhance their online teaching leadership. In fact, the professional development of teachers is not only reflected in the improvement of professional thinking, professional knowledge and professional ability, but also in the fact that teachers become a unique teaching individual. Through the support of big data technology, teachers can recognize the most authentic self, highlight the teaching personality and wisdom, carry out flexible and individualized teaching, and finally realize the individualized professional development of individuals.
Fourth, the development of education big data policy recommendations
The era of big data has arrived. With the increasing awareness of data among governments, enterprises, educational institutions, and the general public, it is possible to foresee the broad prospects for the development and application of big data in the field of education. At present, the Ministry of Education has formulated the "Overall Plan for the Construction of National Education Management Information System". Its primary task is to establish a national education basic database to support the faster, better and more scientific development of education, which marks the Chinese government's The management and application of educational big data is entering a substantive phase. In addition, some companies in the online education market have begun to develop and promote educational data products and services, and the “spot†of explosive growth has emerged.
Compared with the fields of commerce, transportation, environment and medical care, the education field has stronger uniqueness and complexity. There are still many problems in the application of big data technology in education. For example, the limitations of the educational environment are difficult to obtain the learners offline. The behavioral data, the inter-departmental sharing of educational data still has institutional barriers, the mechanism of standardized dynamic collection and real-time updating of educational data has not yet been established, and the leakage of learner privacy data and the risk of improper application are large. The above problems directly affect the application and promotion of educational big data in the field of education and teaching, which makes it difficult for big data technology to play a substantial role in the short term. In addition, with the rapid development of the big data industry in the field of education, many problems and challenges in the security, governance, and operation of educational data will become more and more prominent.
In order to ensure the standardization of education data in the collection, storage, analysis, application, management and other aspects, and promote the sound development of education big data in China, it is urgent to formulate relevant policies for guidance and supervision.
(1) Promulgating the "Guiding Opinions on the Development of Education Big Data Application"
In order to seize the opportunity and promote the healthy and rapid development of the education big data business and industry, it is urgent to formulate corresponding application development guidance. It is recommended that the relevant central ministries and commissions take the lead in formulating the "Guiding Opinions on the Development of Educational Big Data Applications." Its core contents include: increasing support for the application and promotion of educational big data from the national level, upgrading the application of educational big data to a higher strategic level; strengthening the breadth and depth of educational data collection, standardizing the collection process, and increasing education. The intensity of data mining and analysis provides the basis for the development of personalized education and the formulation of science education and teaching management policies; defines the responsibilities of educational authorities at all levels, schools, and related enterprises, highlights the needs of educational development, and focuses on the application of educational data.æœç»æµªè´¹å’Œé¢å工程;鼓励教育大数æ®æŠ€æœ¯äº§å“å¼€æºï¼Œå·¥ä¿¡éƒ¨ç”µä¿¡ç ”究院的《大数æ®ç™½çš®ä¹¦ã€‹åˆ†æžç§°â€œå¤§æ•°æ®æŠ€æœ¯å‘展与开æºè¿åŠ¨çš„结åˆæ˜¯å¤§æ•°æ®æŠ€æœ¯åˆ›æ–°ä¸çš„一个鲜明特点â€ï¼Œæ•™è‚²å¤§æ•°æ®ä¹Ÿåº”èµ°å‘å¼€æºã€å¼€æ”¾;从资金ã€äººæ‰ã€æ”¿ç–ç‰æ–¹é¢ç»™äºˆç›¸åº”支æŒï¼Œå¼•å¯¼æ•™è‚²å¤§æ•°æ®äº§ä¸šå¥åº·æœ‰åºå‘展。
(二)制定《教育大数æ®å®‰å…¨ç®¡ç†åŠžæ³•ã€‹
教育数æ®æ—¢æ˜¯ä¸€ç¬”å®è´µçš„教育资产，åŒæ—¶ä¹Ÿæ¶‰åŠæ•™è‚²è€…å’Œå—教育者的éšç§ï¼Œåº”用ä¿æŠ¤ä¸å½“则会带æ¥ä¸¥é‡çš„安全风险。国家应高度é‡è§†æ•™è‚²æ•°æ®çš„éšç§ä¿æŠ¤ä¸Žå®‰å…¨ç®¡ç†ï¼Œä¸æ–努力采å–更先进ã€å®‰å…¨ç³»æ•°æ›´é«˜çš„措施æ¥ä¿éšœæ•™è‚²æ•°æ®çš„安全，ä¿æŠ¤æ•™è‚²éšç§æ•°æ®ä¸å¤–泄ã€ä¸è¢«æ¶æ„ä½¿ç”¨ã€‚å› æ¤ï¼Œæˆ‘å›½éœ€åŠ å¿«åˆ¶å®šã€Šæ•™è‚²å¤§æ•°æ®å®‰å…¨ç®¡ç†åŠžæ³•ã€‹ï¼Œä»Žä½“制ã€æœºåˆ¶ã€æŠ€æœ¯ã€æ–¹æ³•ç‰å¤šä¸ªå±‚é¢åˆ¶å®šç®¡ç†ç»†åˆ™ï¼Œåˆ‡å®žä¿éšœä¸ªä½“ã€æœºæž„ã€å›½å®¶çš„教育数æ®å®‰å…¨ã€‚
《教育大数æ®å®‰å…¨ç®¡ç†åŠžæ³•ã€‹çš„æ ¸å¿ƒå†…å®¹åº”åŒ…æ‹¬ï¼šå»ºç«‹å¥å…¨æ•™è‚²æ•°æ®å®‰å…¨ç®¡ç†æž¶æž„，包括数æ®ç”Ÿäº§éƒ¨é—¨ã€æ•°æ®ä½¿ç”¨éƒ¨é—¨ã€æ•°æ®ç®¡ç†éƒ¨é—¨ç‰ï¼Œå¹¶æ˜Žç¡®å„部门的安全管ç†èŒè´£;建立教育数æ®èµ„æºçš„ä¿å¯†ç‰çº§ï¼ŒæŒ‰ä¿å¯†ç‰çº§è§„定采å–相关处ç†æŽªæ–½;设计高性能ã€é«˜å¯é ã€ç®€åŒ–管ç†çš„教育大数æ®å˜å‚¨ç³»ç»Ÿï¼Œä¸ºæ•™è‚²äº‘å¹³å°æ供充分的访问性能和数æ®å®‰å…¨ä¿éšœ;建立涵盖教育数æ®å˜å‚¨ã€ä¼ 输ã€ä½¿ç”¨ç‰å¤šä¸ªçŽ¯èŠ‚的多ç§å®‰å…¨ä¿éšœæŽªæ–½;采用数æ®éš”离ã€æ•°æ®åŠ 密ã€ç¬¬ä¸‰æ–¹å®žå认è¯ã€çµæ´»è½¬ç§»ã€å®‰å…¨æ¸…除ã€å®Œæ•´å¤‡ä»½ã€æ—¶é™æ¢å¤ã€è¡Œä¸ºå®¡è®¡ã€å¤–围防护ç‰å¤šç§å®‰å…¨é˜²æŠ¤æŠ€æœ¯ï¼Œè§£å†³æ•™è‚²å¤§æ•°æ®çš„云å˜å‚¨å®‰å…¨é—®é¢˜ã€‚
(三)æˆç«‹å›½å®¶æ•™è‚²å¤§æ•°æ®ç ”究机构
组织教育å¦ã€ç®¡ç†å¦ã€è®¡ç®—机科å¦ã€ç»Ÿè®¡å¦ç‰å¤šå¦ç§‘çš„ç ”ç©¶äººå‘˜æˆç«‹ä¸“门的教育大数æ®ç ”究机构，集ä¸ä¼˜åŠ¿åŠ›é‡ç ´è§£æ•™è‚²å¤§æ•°æ®åº”用推广过程ä¸å˜åœ¨çš„çƒç‚¹ã€éš¾ç‚¹é—®é¢˜ï¼ŒåŒæ—¶ç»“åˆæ•™è‚²å‘展的战略需求，开展å‰çž»æ€§ç ”究，使其æˆä¸ºå›½å®¶æ•™è‚²å¤§æ•°æ®å‘展的智库。
国家教育大数æ®ç ”究机构的主è¦èŒè´£åŒ…æ‹¬ï¼šæ·±å…¥é’»ç ”æ•™è‚²å¤§æ•°æ®æŠ€æœ¯ï¼Œç ´è§£æ•™è‚²æ•°æ®é‡‡é›†ã€å˜å‚¨ã€å¤„ç†ä¸Žåº”用过程ä¸å˜åœ¨çš„关键性技术难题，ç€åŠ›ç ”究教育大数æ®çš„应用模å¼ä¸Žæ–¹æ³•ï¼Œæ›´å¥½åœ°æœåŠ¡ä¸ªä½“ã€å¦æ ¡ã€æ•™è‚²ç®¡ç†æœºæž„以åŠæ•™è‚²ä¼ä¸šçš„å‘展;å‘æŒ¥å¥½å›½å®¶çº§æ™ºåº“çš„åˆ›æ–°ä¸Žå¼•é¢†ä½œç”¨ï¼Œç ”å‘教育大数æ®åº”用的相关技术规范ã€æ ‡å‡†ï¼Œè§„范教育大数æ®äº§ä¸šå‘展;结åˆå›½æƒ…构建特色化的教育大数æ®ç†è®ºä½“系，涵盖ç†è®ºã€æŠ€æœ¯ã€æ¨¡å¼ã€æ–¹æ³•ã€æœºåˆ¶ã€æˆ˜ç•¥ç‰å¤šä¸ªå±‚é¢ï¼ŒåŠªåŠ›ä½¿æˆ‘国在智慧教育与大数æ®ç ”究方é¢ä»Žâ€œéšè¡Œè€…â€å˜ä¸ºâ€œé¢†è·‘者â€;归纳ã€æ炼典型的教育大数æ®åº”用案例与模å¼ï¼Œæ€»ç»“教育大数æ®åº”用ç–略，为全国推广教育大数æ®åº”用ç»éªŒæ供指导。
(å››)æˆç«‹å›½å®¶æ•™è‚²å¤§æ•°æ®æ²»ç†æœºæž„
除了教育大数æ®ç ”究机构，还应æˆç«‹æ•™è‚²å¤§æ•°æ®æ²»ç†æœºæž„。大数æ®æ—¶ä»£ï¼Œæ¯æ—¶æ¯åˆ»éƒ½åœ¨äº§ç”Ÿæµ·é‡çš„ã€å„ç§æ¥æºçš„ã€å„ç§ç±»åž‹çš„教育数æ®ï¼Œå¦‚何ååŒå¤šæ–¹åŠ›é‡è¿›è¡Œé«˜æ•ˆçš„教育数æ®æ²»ç†æ˜¯æ‘†åœ¨æ¯ä½æ•™è‚²ç®¡ç†è€…é¢å‰çš„现实问题。教育数æ®æ²»ç†çš„目的是æå‡æ•™è‚²æ•°æ®è´¨é‡ï¼Œä¿æŠ¤æ•™è‚²æ•°æ®éšç§å®‰å…¨ï¼Œä¿éšœæ•™è‚²æ•°æ®åˆç†åº”用，促进教育数æ®åˆæ³•å…±äº«ã€‚
国家教育大数æ®æ²»ç†æœºæž„的主è¦èŒè´£åŒ…括：出å°æ•™è‚²å¤§æ•°æ®æ²»ç†çš„相关办法，建立完整的教育大数æ®æ²»ç†æ¨¡å¼ï¼Œä¿æŠ¤æ•™è‚²æ•°æ®çš„éšç§å®‰å…¨ï¼ŒæŒ‡å¯¼æ•™è‚²æ•°æ®çš„获å–ã€å½’æ¡£ã€ä¿å˜ã€äº’æ¢ä»¥åŠé‡å¤åˆ©ç”¨;将教育数æ®åº“ã€æ•™è‚²æœåŠ¡å¹³å°ã€æ•™è‚²èµ„æºå¹³å°ç‰äº§ç”Ÿçš„æ•°æ®è¿›è¡Œè§„范化采集与汇èšå…±äº«ï¼Œæž„建综åˆæ€§çš„一体化的教育大数æ®æ²»ç†ä½“ç³»;制定教育基础数æ®é‡‡é›†æ ‡å‡†ï¼Œè®¾ç½®ç›¸å…³è¦æ±‚，形æˆæ¸…æ™°çš„æ•°æ®æ²»ç†æœºåˆ¶å’Œæµç¨‹;制定教育数æ®è´¨é‡æ ‡å‡†å’Œæ•™è‚²æ•°æ®ç®¡ç†æˆ˜ç•¥ï¼Œç¡®ç«‹æ•™è‚²æ•°æ®é›†çš„归档和长期ä¿å˜çš„机制与方法;建立教育大数æ®å¼€æ”¾å¹³å°ï¼Œå‘挥ä¼ä¸šã€ä¸ªä½“ã€æ•™è‚²æœºæž„ç‰ä¸»ä½“çš„åˆ›é€ æ€§ï¼Œå¼€å‘大é‡ç‰¹è‰²åŒ–的教育应用，让社会力é‡å…±åŒå‚与教育数æ®çš„æ²»ç†ä¸Žåˆ›æ–°åº”用;监控教育大数æ®çš„应用状况，对éžæ³•åº”用ã€ä¾µçŠ¯ç”¨æˆ·éšç§ä¸Žå›½å®¶å®‰å…¨çš„å•ä½å’Œä¸ªäººè¿›è¡Œå¤„罚，建立教育大数æ®åˆæ³•ã€åˆç†åº”用的良好环境。
(五)é¢å¸ƒâ€œæ•™è‚²æ•°æ®è¿è¥å•†â€ç‰Œç…§
从国家层é¢æ¥çœ‹ï¼Œæ•™è‚²æ•°æ®çš„安全性ä¸äºšäºŽé‡‘èžæ•°æ®ã€‚教育数æ®çš„开放程度ã€å¼€æ”¾èŒƒå›´ã€å¼€æ”¾å¯¹è±¡éƒ½åº”进行深入的论è¯ï¼Œä»¥ç¡®ä¿æ•™è‚²æ•°æ®çš„使用是åˆæ³•çš„，是有利于教育创新å‘展的，而éžé˜»ç¢ã€ç ´åã€å¨èƒåˆ°å›½å®¶å®‰å…¨ã€‚为æ¤ï¼Œæ”¿åºœéƒ¨é—¨åº”对教育数æ®åŠ 强监管，å‚考通信领域é¢å‘è¿è¥å•†ç‰Œç…§çš„æ–¹å¼åˆ¶å®šæ•™è‚²æ•°æ®è¿è¥å•†å‡†å…¥æ ‡å‡†ï¼Œé¢å‘è¿è¥ç‰Œç…§ã€‚建议由教育部ã€å‘改委ã€å•†åŠ¡éƒ¨ç‰å¤šä¸ªéƒ¨å§”è”åˆåˆ¶å®šç›¸å…³è§„定，对国内教育数æ®åº”用行业实施æ£å¼çš„监管，设置教育数æ®ä½¿ç”¨çš„æ ‡å‡†å’Œé—¨æ§›ä»¥åŠç›¸å…³ç”³è¯·èµ„æ ¼ä¸ŽåŠžæ³•ï¼Œé¢å¸ƒâ€œæ•™è‚²æ•°æ®è¿è¥å•†â€ç‰Œç…§ï¼Œéžå®˜æ–¹æœºæž„进行教育数æ®ç ”究与使用，需è¦æŒ‰æ¡ä»¶æ»¡è¶³ç›¸å…³è§„定并å–得相关许å¯ï¼Œæ‰èƒ½èŽ·å¾—教育数æ®çš„使用æƒåˆ©ã€‚
(å…)åŠ å¿«å»ºè®¾æ•™è‚²å¤§æ•°æ®äº§ä¸šåŸºåœ°
以基于海é‡æ•°æ®èµ„æºçš„挖掘和应用催生的大数æ®äº§ä¸šä¸ºåŸºç¡€çš„æ–°ç»æµŽæ˜¯ç¤¾ä¼šç»æµŽå‘展到一定阶段的必然选择。教育大数æ®äº§ä¸šä½œä¸ºå¤§æ•°æ®äº§ä¸šå®¶æ—的一部分，åŒæ ·æ˜¯ç¤¾ä¼šç»æµŽä¸Žæ•™è‚²å‘展的必然选择。当å‰ï¼ŒåŠ 快教育大数æ®äº§ä¸šåŸºåœ°å»ºè®¾ï¼Œæ—¢æ˜¯æ™ºæ…§æ•™è‚²äº‹ä¸šå‘展的需è¦ï¼Œä¹Ÿæ˜¯å¤§æ•°æ®åœ¨æ•™è‚²é¢†åŸŸå‘挥å˜é©ä½œç”¨çš„需è¦ã€‚
ä¸ºåŠ å¿«å»ºè®¾æ•™è‚²å¤§æ•°æ®äº§ä¸šåŸºåœ°ï¼Œæå‡ºå¦‚ä¸‹å‡ ç‚¹å»ºè®®ï¼šå¤šæ¸ é“å¢žåŠ æ•™è‚²å¤§æ•°æ®äº§ä¸šåŸºåœ°çš„投入;æ‰“é€ é€‚åº”æ•™è‚²å¤§æ•°æ®äº§ä¸šå‘展的é…套软硬件环境;åŠ å¿«æˆç«‹æ•™è‚²å¤§æ•°æ®ç ”究机构，产生一批引领型ä¼ä¸š;æ‰“é€ æ•™è‚²å¤§æ•°æ®äº§ä¸šçš„基础é…套设施ã€åŽå°ä¿éšœï¼ŒåŠ›äº‰æ‰“通全产业链æ¡;领先ä¼ä¸šè¿›è¡Œå‰æ²¿æŠ€æœ¯åˆ›æ–°ï¼Œåˆ›æ–°æˆæžœé€šè¿‡å¼€æºå¾—到ä¸æ–完善并å‘全社会è¾å°„，集众人之智实现教育大数æ®äº§å“与æœåŠ¡çš„æŒç»åˆ›æ–°å‘展;æˆç«‹æ•™è‚²å¤§æ•°æ®äº§ä¸šè”盟和å›åŒºï¼Œç»¼åˆæŠ•èµ„机构ã€é«˜æ ¡ä¸Žç§‘ç ”æœºæž„ã€æ”¿åºœéƒ¨é—¨çš„力é‡ï¼Œé‚€è¯·å›½å†…外知å专家å¦è€…å’Œäº§ä¸šäººå£«å®šæœŸç ”è®¨ã€æŒ‡å¯¼äº§ä¸šåŸºåœ°å‘展，努力将基地建设æˆä¸ºå›½å†…外知å的教育大数æ®äº§ä¸šé›†ç¾¤åœ°å’Œå¼•é¢†æ•™è‚²å¤§æ•°æ®äº§ä¸šå‘å±•çš„æ ¸å¿ƒåœ°å¸¦ã€‚
references:
[1]陈ç³ï¼Œé™ˆè€€åŽ.以信æ¯åŒ–带动教育现代化路径探æž[J].æ•™è‚²ç ”ç©¶ï¼Œ2013，(11)：114-118.
[2][14]陈ç³.促进深层å¦ä¹ 的网络å¦ä¹ 资æºå»ºè®¾ç ”究[J].ç”µåŒ–æ•™è‚²ç ”ç©¶ï¼Œ2011，(12)：69-75.
[3]æ¨æ·¼æ·‡ï¼Œå™çº³æ–°ï¼ŒæŸ´åŽ.大数æ®æ—¶ä»£æ•™è‚²æ¨¡å¼çš„ç ”ç©¶[J].计算机工程与科å¦ï¼Œ2014，36(1)：272-273.
[4]喻长志.大数æ®æ—¶ä»£æ•™è‚²çš„å¯èƒ½è½¬å‘[J].江淮论å›ï¼Œ2013，(4)：34.
[5]ç¥æ™ºåºï¼Œæ²ˆå¾·æ¢….基于大数æ®çš„æ•™è‚²æŠ€æœ¯ç ”ç©¶æ–°èŒƒå¼[J].ç”µåŒ–æ•™è‚²ç ”ç©¶ï¼Œ2013，(10)：5-13.
[6]å¾é¹ï¼ŒçŽ‹ä»¥å®ï¼Œåˆ˜è‰³åŽï¼Œå¼ æµ·.大数æ®è§†è§’分æžå¦ä¹ å˜é©â€”—美国《通过教育数æ®æŒ–掘和å¦ä¹ 分æžä¿ƒè¿›æ•™ä¸Žå¦ã€‹æŠ¥å‘Šè§£è¯»åŠå¯ç¤º[J].远程教育æ‚志，2013，(6)：11-17.
[7]顾å°æ¸…ï¼Œå¼ è¿›è‰¯ï¼Œè”¡æ…§è‹±.å¦ä¹ 分æžï¼šæ£åœ¨æµ®çŽ°ä¸çš„æ•°æ®æŠ€æœ¯[J].远程教育æ‚志，2012，(1)：18-25.
[8]è‘›é“å‡¯ï¼Œå¼ å°‘åˆšï¼Œé顺平.教育数æ®æŒ–掘：方法与应用[M].北京：教育科å¦å‡ºç‰ˆç¤¾ï¼Œ2012.
[9]维基百科.Adaptive Learning[EB/OL].[2014-07-01].http：//en.wikipedia.org/wiki/Adaptive_learning.
[10]陶雪娇，胡晓峰，刘洋.大数æ®ç ”究综述[J].系统仿真å¦æŠ¥ï¼Œ2013，(1)：142-146.
[11]é顺平.å¦ä¹ 分æžæŠ€æœ¯ï¼šæŒ–掘大数æ®æ—¶ä»£ä¸‹æ•™è‚²æ•°æ®çš„价值[J].现代教育技术，2013，(2)：5-11.
[12]æŽå©·ï¼Œå‚…钢善.国内外教育数æ®æŒ–æŽ˜ç ”ç©¶çŽ°çŠ¶åŠè¶‹åŠ¿åˆ†æž[J].现代教育技术，2010，(10)：21-25.
[13]陈ç³.ä¸å›½é«˜æ ¡æ•™è‚²ä¿¡æ¯åŒ–å‘展战略与路径选择[J].æ•™è‚²ç ”ç©¶ï¼Œ2012，(4)：50-56.
[15]Mayer-Schonberger,V.,Cukier,K..Big Data:A Revolution That Will Transform How We Live,Work,and Think[M].New York:Houghton Mifflin Harcourt,2013.
[16]陆璟.大数æ®åŠå…¶åœ¨æ•™è‚²ä¸çš„应用[J].ä¸Šæµ·æ•™è‚²ç§‘ç ”ï¼Œ2013，(9)：5-8.
[17]Schleicher,A..Big Data and PISA[EB/OL].[2014-04-25].http://
[18]刘é›æ½œï¼Œæ¨çŽ°æ°‘.大数æ®æ—¶ä»£åŒºåŸŸæ•™è‚²å‡è¡¡å‘展新æ€è·¯[J].ç”µåŒ–æ•™è‚²ç ”ç©¶ï¼Œ2014，(5)：11-14.
[19]通信世界网.æ•°æ®æŒ–掘人工智能使教育定制化[EB/OL].[2014-05-01].http：//
[20]Gall Button.大数æ®è¿›æ ¡å›[EB/OL].[2014-05-01].http：//weibo.com/2357403745/AFfeR4HCu?mod=weibotime.
[21]王è¿æ¦.我国数å—æ ¡å›å»ºè®¾ç ”究综述[J].çŽ°ä»£è¿œç¨‹æ•™è‚²ç ”ç©¶ï¼Œ2011，(4)：39-50.
[22]æ¨æ°¸æ–Œ.æ•°æ®æŒ–掘技术在教育ä¸çš„åº”ç”¨ç ”ç©¶[J].计算机科å¦ï¼Œ2006，33(12)：284-286.
[23]Pistilli,MD,Arnold,KE.In Practice:Purdue Signals:Mining Real-time Academic Data to Enhance Student Success[J].About Campus,2010,15(3):22-24.
[24]ç¿Ÿåš.教育å‡è¡¡å‘展：ç†è®ºã€æŒ‡æ ‡åŠæµ‹ç®—方法[J].æ•™è‚²ç ”ç©¶ï¼Œ2006，(3)：16-28.
Kitchen Shears,Chef Knife Set,Black Kitchen Knife Set,Knife Block Set
Yangjiang Teammade Industrial Co., Ltd , https://www.yadokitchen.com