Key issues concerning the development of intelligent education

Sannyuya Liu, Shengyingjie Liu, Jianwen Sun, Xiaoxuan Shen and Zhi Liu

A global wave of interest in the new generation of intelligent technology has the potential to accelerate intelligent education development and become a new driving force for educational innovation and transformation. Currently, research and development of intelligent education face numerous challenges due to limitations in terms of intelligent technology maturity, black-box nature of algorithms, human-computer integration and human-computer mutual trust. This article expounds on key issues in relation to the above limitations and explores directions for future research and development of intelligent education, namely constructing new infrastructure for education, making breakthroughs in developing generic technologies, refining ethical norms, enhancing interdisciplinarity, and strengthening multi-actor collaboration.

Keywords: intelligent education; artificial intelligence; technology empowerment; generalization ability; black box; cognitive mechanism; human-computer integration; human-computer mutual trust

Towards a prediction model of learning performance: informed by learning behavior big data analytics

Hang Hu, Shuang Du, Jiarou Liang and Zhonglin Kang

With the development of educational big data analytics, learning prediction has become an important component of learning analytics. Nevertheless, there is a dearth of learning prediction research based on learning behavior data in different scenarios. Using log data of 823 university students collected in two settings: their online learning setting and daily life setting (using campus ID cards for consumption purposes and book-borrowing in the university library), this study creates indicators for online learning behavior, early-rising behavior, book-borrowing behavior and learning performance prediction. Five machine learning models are employed to analyze learning performance prediction, with the additional use of Boosting and Bagging to improve the accuracy of the prediction model. The predictability of the proposed model is also compared with that of both the Artificial Neural Network model and the Deep Neural Network model. Meanwhile, a learning behavior prediction diagnosis model is constructed by using the classification rule set which is created by combining the decision tree and the rule model. Findings show that multi-scenario behavior performance indicators have strong predictive capabilities while the Deep Neural Network model has the highest prediction accuracy (82%) but is most time-consuming. The model based on the rule set is highly accurate, readable and operable and may be conducive to making accurate teaching interventions and resource recommendations.

Keywords: university students; learning behavior; multiple scenarios; learning performance; prediction model; machine learning; decision tree; neural network

Educational data availability on Chinese local government platforms: status quo, problems and solutions

Qing Li and Hailan Wang

Against the backdrop of the booming Internet and the rising open data movement, governments across the world have begun to include open data in their big data implementation strategies. China is also beginning to follow suit. The opening of educational data can contribute to government transparency, effective decision-making in educational institutions, and social engagement. This study accesses 14 provincial government platforms and 15 prefecture-level government platforms in China to investigate their functions and educational data availability. Discussion is informed by relevant literature as well as policies, laws and regulations. The findings are then compared with best practice in other countries. Gaps are identified between China and other countries in terms of data opening permission and privacy protection; platform function and service quality; data collection, release and update; data value and quality, and data reuse. Suggestions are also made concerning relative areas.

Keywords: educational data; open data; information disclosure; government data; regional data; open platform; government administration

The evolution of portfolio in higher education: past, present and future

Orna Farrell

This article traces the evolution of the concept of portfolio from the renaissance to the present day. The meaning of portfolio evolved from its origins as a case for holding loose papers to other contexts such as finance, government and education. Portfolios evolved from paper to electronic, from local network to the World Wide Web. The decade from 2000-2010 was a period when technology became part of mainstream society and educational technology became part of mainstream higher education, and portfolios spread around the world. As a shift in focus occurred in eportfolio research and practice in the last decade, there was more emphasis on pedagogy and student learning and less focus on digital technology which had become ubiquitous. One of the key lessons from the story of eportfolio adoption is that educators and institutions should adopt a critical perspective to new educational technologies and approaches. Finally, the history of portfolio in higher education shows that the higher education system will continue to gradually evolve, incorporating concepts, technology and approaches that are compatible rather than transformative.

Keywords: eportfolio; portfolio; history of edtech; edtech; higher education

An empirical study on lifelong learning for the disabled

Ran Bai, Hao Xie and Yusen Hu

In 2019, the Fourth International Conference on Learning Cities adopted inclusion as a fundamental principle for lifelong learning and sustainable cities. As a typical vulnerable group, the disabled should be a major target group to be catered for in constructing learning cities. The development of Internet technology has provided a new opportunity for the disabled to participate in lifelong education. This study administers a questionnaire survey among 794 disabled people to investigate the current situation of their lifelong learning and continuing education. Findings cover several aspects. First, the participants are found to recognize the importance of lifelong learning but need to learn more about it. They participate in lifelong learning chiefly for physical and psychological development. There is diversity both in terms of form and content, albeit their preference for traditional modes of learning over online learning. Furthermore, they are relatively less proficient in using ICT tools in addition to lack of time as well as interesting and affordable courses. Finally, it is found that individual differences were also factors influencing their lifelong learning. Implications from the above findings are then discussed.

Keywords: disabled people; disabled adult; education for the disabled; lifelong learning; inclusion; learning motivation; learning mode; ICT

（英文目次、摘要譯者：肖俊洪）