Based on the extensive penetration of cloud, Internet of Things, database technology, artificial intelligence and virtual reality in the field of education, the era of big data is giving birth to a series of new educational changes. How does the existing teaching and research mechanism adapt to the future dominated by big data? How to innovate school-based teaching and research ideas and strategies based on the application of big data? How to find the best transformation path of school-based teaching and research based on big data platform? This is exactly the question that this paper tries to answer.

First, traditional teaching and research lack the limitations of big data.

We live in an era surrounded by geometric explosion data, and all our actions are generating massive data, which is called "big data". [1]2 The "big" of big data is not only the "big capacity", but also the greater significance is that with the help of cloud technology, through the screening, integration and analysis of massive data, new problems can be solved and new values created. In the era of big data, the traditional form of school-based teaching and research has lagged behind the development of the times.

First, the management of teaching and research based on administrative orders can no longer adapt to the new trend of teaching and research development. Traditional school-based teaching and research activities are characterized by too much administrative guidance, too strong task drive, less interactive generation and lack of process data. Their teaching and research forms are single, closed, lagging and static. Teaching and research activities are mostly "prescriptive actions", and most schools follow the model of "the school arranges the teaching and research plan at the beginning-the teaching and research group arranges and implements it as planned-organizes teaching demonstrations or discussions in stages-and finally summarizes the teaching and research in the department". [2] In such a unified pace of "rapid March", it is difficult for teaching and research to focus on the detailed study of real and difficult problems in the classroom, and discussion activities rarely touch the depth of exploring the laws of general disciplines. School-based teaching and research lacks real research process, and case tracking lacks process data. Without the support of process and actual data, the effectiveness of teaching and research is greatly reduced, and it is difficult to effectively broaden the horizons of teaching and research. This form of teaching and research activities is obviously insufficient in aggregation, openness, innovation and dynamics, and its formalization and inefficiency are prominent.

Second, the teaching and research method based on experience can no longer adapt to the new situation of teaching and research development. At present, most schools' understanding and research on big data is still in its infancy, and the exploration and practice of big data in the field of education has just begun. There is a general lack of in-depth understanding and operational strategies on how to truly apply big data to teaching and scientific research and feed it back to education. On the one hand, due to the lack of "bottom-up awareness of theme confirmation" in traditional teaching and research, limited by space, time and technical equipment, the teaching and research content can not directly arouse the internal needs of teachers' teaching and research, can not directly produce radiation and benefits to their classrooms, and is difficult to attract teachers' deep participation. On the other hand, in the face of the ever-changing teaching and research situation in the new era, more schools have not taken the initiative to keep pace with the times, unable to provide teachers with a teaching and research platform for free thinking collision (such as providing a discussion platform before and after activities), unable to provide them with timely and sufficient data support and technical convenience needed for educational research, and lacking sufficient cohesion and attraction for teachers' diversified cooperation and in-depth expansion of teaching and research. Such teaching and research will inevitably fail to show its due value and vitality in the era of big data.

Second, the transformation strategy of school-based teaching and research in the era of big data

Facing the era of big data, facing the new educational situation of cloud computing, cloud storage, cloud education, cloud classroom and cloud teaching and research with "mobile Internet and tablet touch technology" as the core, how should primary and secondary schools respond to this unprecedented educational reform? What are the good teaching and research strategies and application paradigms?

(A) the change of the core perspective: from focusing on "teaching" to focusing on "learning"

"Teacher-centered" or "student-centered" once became a hot topic in teaching and research. Based on "who", it reflects the renewal of a teaching concept and shows a choice of teaching behavior. In traditional teaching and research, the evaluation of students is often the thinnest-most participants can only imagine students' experiences according to their own experiences. The disadvantage of this scratch-off evaluation stems from the lack of an important factor-there are not enough data sources to provide academic analysis and empirical evidence.

The arrival of the era of big data can effectively change this lack. With the help of video transmission, data collection, peer-to-peer terminal, cloud storage server and personalized data analysis software, the problem of lack of data sources can be solved from the technical level, and the data can be comprehensively processed and analyzed, which can accurately quantify and show students' personalized feelings. Students' real needs and immediate mentality in the process of teaching activities can also be readable, visible and quantifiable through technical transformation and dynamic integration analysis. This provides strong technical support and interpretation guarantee for the transformation of teaching and research perspective from focusing on "teaching" to focusing on "learning", and provides more vivid materials for the research and prediction of learning situations. We can even use streaming video and data analysis to help teachers track students' real-time learning, so as to make corresponding teaching plans and adjust strategies according to students' ability level, and better carry out targeted personalized learning research.

The application of big data gives the classroom a new perspective of observation and teaching discussion a new perspective. It has become possible to change teachers' teaching behavior from the perspective of students' learning analysis and prediction. When technology can help us understand the needs of each student, is the concept of "teaching students in accordance with their aptitude" that has lasted for more than two thousand years closer to us?

(b) Normal method transformation: from repeated experience to empirical data.

Traditional school-based teaching and research is often empirical. We always subjectively assume that some educational factors are important to students and which variables have an impact on the classroom, and then verify the reliability of these subjective experiences through repeated practice according to our own judgment. This experience-oriented traditional teaching and research often has subjective, speculative and enlightening defects, and it is often prone to the situation that "the problem is not at the grassroots level, the goal is broad, and the evidence is relatively lacking". After the research, there is no qualitative improvement in understanding and solving the problems compared with before the research. The reason is that there is no in-depth questioning and data investigation before the study, insufficient data comparison and logical analysis in the study, and insufficient deductive analysis and practical verification after the study.

The arrival of the era of big data has found the basis and direction for effectively solving the stubborn problem of repeated experience teaching and research. According to the collection, arrangement and analysis of experimental data, the theme of teaching and research is effectively established, so that the research directly points to the solution of practical problems; Relying on the platform of "cloud teaching and research, cloud management", the data that could not be collected and analyzed in the past were given the possibility of obtaining through new technical means, which provided technical support for effectively exploring problems and demonstrating topics. This kind of teaching and research based on data argumentation is more scientific, logical and convincing. As Dr. Zhong Wei said in the book "Education is quietly undergoing a revolution": "Education will no longer be a subject of social science and moral conscience inherited by ideas and experience after economics, and education in the era of big data will become a real empirical science." [1]3 Clear target monitoring, massive data support, clear process case data, powerful data analysis and demonstration, and a teaching and research innovation platform closely integrated with it make teaching and research more innovative.

(C) the change of practice: from the study of scattered problems to the study of a series of projects.

"Guiding" and task-driven teaching and research are common in primary and secondary schools in China. Its passive acceptance of research mentality, conclusion-oriented research method, and scattered point activities instead of the discussion mode of series of research practices make school-based teaching and research difficult to be grounded, which eventually leads to low credibility of teaching and research results and difficult to popularize. Teaching and research in the new era must change from formal, superficial and decentralized teaching and research forms to thematic, serialized, thematic and project-based teaching and research, which is also determined by the characteristics of teaching and research in the era of big data.

In the era of big data, because the teaching platform, teaching and research platform and management platform have been effectively connected, all levels and series of data can be shared on the "cloud platform" of big education, and big data technology will be skillfully applied to all aspects of classroom, teaching and research. Whether it is top-down data calling or bottom-up data collection, it has become or can become the normal state of primary and secondary education teaching and research. The innovation and transformation of technical means have created conditions for the storage, integration and analysis of educational big data. "Wake up the data to talk", especially in the era of big data, researchers began to consciously and consciously collect, transform and use data as a serial, periodic and comparable normal research. This kind of screened data will become the most powerful display of thematic teaching and research and project research, and will also bring sudden changes in teaching and research ideas to those teachers who were originally troubled by technical or conditional restrictions and could not easily obtain research data, which will eventually be realized.

Third, the transformation path of school-based teaching and research in the era of big data

(a) "Cloud classroom" research: technology and data better serve "learning"

The new technology is around, whether you use it or not, it is there. Xinghai Primary School in Suzhou Industrial Park brought "ipad into the classroom" and promoted teaching and scientific research with digital technology, which provided a new perspective for us to study the application of big data. In 20 14, the school started experimental teaching with mobile network as the platform and ipad as the terminal. Through the application of new technology, a "cloud classroom" with students as the main body has been constructed, which has greatly changed the traditional way of "teaching" and "learning". The core of cloud classroom technology support is "cloud computing". It is a computing method, which provides users with a large amount of information and resources on demand through unified management and scheduling of a large number of network connections. This holographic service network is called "Cloud". "Cloud" is like a professional "information ATM". Its powerful information technology and extremely rich three-dimensional data resources have built a multi-dimensional interactive "cloud platform" for students' learning, teachers' teaching and team research. [3]

Based on the research on the reform of teaching and research methods in the era of big data, on the basis of full investigation, demonstration and research and development, the school has built an independent cloud storage server for digital learning and research of teachers and students. Each ipad is equipped with a "network disk wizard" for cloud storage and sharing, in which students and teachers can build a separate storage space. The courseware made by each teacher and the experimental data collected can be uploaded to the server in the first time, and the teachers and students of the whole school can download the required resources in the first time. Teachers in various subjects can also cooperate in data uploading and downloading, data storage and analysis with the help of wireless platforms and application software. Many large and small teaching and research activities are carried out with the support of a powerful data platform. Online teaching and research has enabled more teachers to enhance their teaching and research motivation. For example, after the ipad is connected to the English classroom, students can record their own videos of reading and singing at home and upload them to the "network disk wizard", which provides first-hand research and analysis materials for teachers to know the students' learning status and learning effectiveness in real time. Vivid data makes the tracking and prediction of teaching become the norm. [4]

For another example, the "Three-Network Wisdom Pan-Education Cloud Platform" independently developed by Asia Education Network is a kind of public wisdom cloud of "Three Networks Convergence and Ubiquitous Learning". It uses new technologies such as cloud computing, Internet of Things and virtualization to upgrade campus network and metropolitan area network. The new model of "educational cloud+interactive TV+e-bag" initiated by it has opened a new era of educational informatization and realized the sharing of educational resources in many schools and places in an all-round and large-scale way. [5] 10- 1 1

(B) "empirical research": strengthen data argumentation and explore the "universal" law.

In recent years, the exploration of teaching and research based on data argumentation has been continuously carried out, and there have been microteaching, fragment teaching and research, and theme teaching and research. Most of these teaching and research models adopt the way of "empirical research". They are all research methods to obtain objective data through a large number of observations, experiments and investigations on the research objects, and to summarize the essential attributes and development laws of things from individual to general. These teaching and research models are based on problem research, with teaching cases as the carrier, data analysis as the means, and show teaching and research work in a micro, fragmented, pre-positioned and thematic way. Relying on the law of data exploration, the results of teaching and research are more clear, dominant and effective.

In recent years, Jing 'an District of Shanghai began to implement the practical project of "Cultivating Social Emotional Ability" in 7 kindergartens and 9 primary schools. After more than four years' exploration and improvement by nearly 100 teachers, the course of "Cultivating Social Emotional Ability" in primary schools in Jing 'an District has gradually matured, and one of its research methods is "empirical research". Taking "social emotion" as the theme, they explored a data collection method that relied on "data terminal" to record the performance of every student, every class and every link. For example, in a class with six links, most of the time the students follow the teacher closely, but in a certain link, most of the students stay far longer than the teacher. This reminds us that this link needs to be studied and adjusted. Perhaps this part of the content is very attractive to students, or it may be difficult, and it needs more time to read and digest. This kind of teaching and research method with the help of big data is also suitable for us to capture the microscopic research of students more effectively in the classroom. It can be said that the arrival of the era of big data makes it possible to track every data, thus making it possible to study "human nature". For educational researchers, we will be closer to discovering real students than ever before.

(c) "Project research": supporting evaluation with evidence and promoting teaching and research with projects.

In 20 14, Suzhou Education Bureau set up five major educational reform projects, such as "comprehensive evaluation reform of compulsory education quality", and selected 50 schools with distinctive characteristics from the whole city to form a project school community to promote the implementation of project research. In the research process, each project school effectively upgraded the campus network with the help of new technologies such as cloud computing, Internet of Things and virtualization, and vigorously integrated cloud technology and Internet of Things, and made necessary hardware configuration and software development for uploading, storing, integrating and analyzing all-round and personalized process data and research resources. Then, under strict process management, a series of special research and project practice demonstrations based on platforms, cases and data were carried out. At present, the project research is progressing smoothly and has achieved gratifying results. Taking Xinghai Primary School in Suzhou Industrial Park as an example, the school replaced "three good" with "top ten" and launched a multi-evaluation system of "top ten Xinghaiwa", which initiated the practice and research of "comprehensive evaluation reform of compulsory education quality" in Suzhou. In project research, comprehensive resources and personalized data collection and analysis are the basis of project promotion, and sharing research resources and educational achievements in the research process is the key to successful research. In order to effectively integrate resources and visualize data, Xinghai Primary School in Suzhou Industrial Park has launched personalized practice cases such as "Xinghaiwa" self-declaration, "Four-leaf clover" medal, collection of golden ideas and series recruitment of ordinary citizens, which broadened the evaluation system and enriched the evaluation data. Matching with Xinghaiwa's evaluation system, Xinghai Primary School in Suzhou Industrial Park has also innovated multiple evaluation support systems such as the "Four-leaf clover" citizen practice center, and started to develop an online tracking platform for Xinghai teachers and students' growth files, fully supporting the green evaluation system, and eliminating information islands among schools, families and society through seamless connection of open and shared "cloud" resource platforms, which are realized through terminals such as computers, mobile phones, televisions and tablets. [5] The preliminary exploration of the pilot project of Xinghai Primary School in Suzhou Industrial Park on 281-282 shows that data demonstration makes teaching and research more accurate and scientific, and "supporting evaluation with evidence and promoting teaching and research with projects" has become a feasible way for school-based teaching and research.