The development of new generation artificial intelligence technologies such as deep learning, knowledge graphs, and reinforcement learning is driving the “Internet + Education” into a new era of “Smart Education.” Knowledge graphs, as a core driving force for the advancement of artificial intelligence, provide new empowering capabilities for education and teaching in the era of Education Informatization 2.0. From the perspective of research paradigms in artificial intelligence, knowledge graphs represent the evolution and development of the symbolic research paradigm in the era of big data and artificial intelligence. From the development stages of artificial intelligence, knowledge graphs serve as an important foundation for the progression from “perceptual intelligence” to “cognitive intelligence.” Therefore, from the perspective of “AI +,” educational knowledge graphs have broad application prospects in areas such as intelligent processing of educational big data, semantic aggregation of teaching resources, optimization of smart teaching, construction of learner profiling models, adaptive learning diagnostics, personalized learning recommendations, and intelligent educational robots.

1. Introduction to Knowledge Graphs

A knowledge graph is a large-scale semantic network that represents a semantic form of the real world. Entities are represented as nodes, and the attributes of entities and the relationships between entities are represented as edges, forming a network structure. This structured form is recognizable by humans and friendly to machines, facilitating machine understanding. The large-scale concepts, attributes, and relationships among entities in the graph provide rich semantic information and associative information. The unique characteristics of the graph enable various operations and applications. Once constructed, it can also serve as background knowledge for downstream applications. The concept of the knowledge graph was initially proposed by Google to improve search engine capabilities and enhance search quality. Due to the unique characteristics of knowledge graphs, they can play an important role in various aspects of artificial intelligence.

Applications of Educational Knowledge Graphs in the Context of AI+

2. Connotation and Classification of Educational Knowledge Graphs

(1) Connotation of Educational Knowledge Graphs

Currently, there is no unified understanding of the concept of educational knowledge graphs in academia, and different researchers have interpreted it from various research perspectives. We should recognize educational knowledge graphs from multiple dimensions: from the perspective of knowledge modeling, educational knowledge graphs are a method for modeling the ontology of subject knowledge; from the perspective of resource management, educational knowledge graphs semantically organize resources and their relationships in the education field in the form of a “graph”; from the perspective of knowledge navigation, educational knowledge graphs can generate personalized learning paths aimed at learning objectives with the support of technologies such as big data and artificial intelligence; from the perspective of learning cognition, a cognitive schema of learners can be formed based on knowledge graphs combined with learners’ knowledge mastery status information; from the perspective of knowledge bases, educational knowledge graphs are a structured semantic knowledge base that stores knowledge in the education field in a way that can be “understood” by computers.

(2) Classification of Educational Knowledge Graphs

Educational knowledge graphs can be divided into the following two categories: (1) Static Knowledge Graph (SKG), which consists of elements involved in the teaching process as entity nodes and the logical relationships between teaching elements as edges, forming a semantic network. Teaching elements can refer to knowledge points in the subject (concepts, formulas, theorems, principles, etc.) or broadly refer to textbooks, courses, teaching resources, knowledge themes, knowledge units, teaching objectives, teaching questions, teaching participants, teaching situations, etc.; (2) Dynamic Reason Graph (DRG), which represents teaching events or activities as objects and uses logical relationships (sequence, causality, reversal, conditions, hierarchy, composition, etc.) as edges to form a multi-relational graph. The following images show static knowledge graphs focused on knowledge points and dynamic reason graphs focused on activities.

Static Knowledge Graph Focused on Knowledge Points

Dynamic Reason Graph Focused on Activities

3. Applications of Educational Knowledge Graphs in the Context of AI+

(1) Supporting Intelligent Processing of Educational Big Data

Educational big data is the foundation for AI applications in education. The analysis and mining of educational big data is an extraction process from “data” to “knowledge.” Through knowledge graphs, we can integrate the vast and complex big data in the education field into a semantic knowledge network, solving the challenges of data aggregation and integration, and optimizing the data value extraction process, as shown in the following image.

Application of Knowledge Graphs in Intelligent Processing of Educational Big Data

From the perspective of data aggregation and integration, the low value density and sparse nature of educational big data require the integration of multi-source heterogeneous educational data to ensure the accuracy of analysis results. However, the current educational big data faces serious issues such as a lack of unified standards and norms in the aggregation and integration process, difficulties in dynamic changes of data models, and challenges in semantic integration of multi-source heterogeneous data. Problems like “data islands” and “data silos” remain severe. Therefore, there is an urgent need for a flexible, scalable, and intelligent adaptive data model to achieve multi-dimensional deep integration of existing data. Knowledge graphs, as a lightweight data model with semantic associations and dynamic scalability, can, to some extent, achieve unified modeling and management of multi-source heterogeneous data. First, subject experts and knowledge engineers need to define the standard model of the knowledge graph from three dimensions: teaching, management, and research. Second, big data engineers need to map massive teaching log data, teaching resource data, and learning behavior and evaluation data to the standard model. Third, through knowledge validation techniques, entities from different data sources need to be integrated. Finally, entity linking technology should be used to associate data with various resource and knowledge bases.

(2) Supporting Semantic Aggregation of Teaching Resources

In recent years, with the emergence of ubiquitous learning environments and open educational resources such as MOOCs, SPOCs, and micro-courses, the way knowledge is acquired has shown cross-end, cross-source, and cross-modal characteristics. Learning resources face serious issues such as decentralization, knowledge fragmentation, difficulties in sharing, and lack of associations, making semantic aggregation of learning resources a hot topic in educational technology research.

Knowledge graphs, with their ability for semantic association and intelligent organization and aggregation, provide new ideas for machines to understand complex learning resources and construct knowledge semantic networks. They are one of the key technologies for organizing, representing, and managing massive educational resources and achieving integration of educational resources.

(3) Enhancing the Efficiency of Smart Teaching

Supporting “teaching based on learning” where teaching objectives serve as the starting point and guiding compass of the teaching process. Their precise positioning determines the effectiveness of teaching design and the teaching process. Artificial intelligence technologies represented by knowledge graphs provide new technical means for the precise positioning of teaching objectives, mainly reflected in: (1) Based on subject knowledge graphs, it can accurately detect students’ mastery status of various knowledge points. With the help of educational big data collection technologies, intelligent learning systems can record students’ learning trajectories in various aspects such as assignments, exercises, exams, and Q&A. Combined with learning analytics technologies, students’ mastery levels of knowledge points can be visualized in the form of knowledge graphs, accurately identifying students’ learning weaknesses and fragile knowledge points. (2) Combining knowledge tracking technology and relevant educational theories, it can dynamically predict changes in students’ mastery of various knowledge points. Each student’s learning status, progress, and knowledge level are unique and dynamically changing, requiring the assistance of artificial intelligence technologies for dynamic prediction.

(4) Empowering the Construction of Learner Profiling Models

User profiling is a modeling method based on user behavior big data, integrating techniques such as text mining, sentiment analysis, knowledge extraction, and data visualization to describe users’ multi-dimensional characteristics in a labeled information model. The essence is to “tag” the subject of the profile. Learner profiling is a special form of user profiling, mainly used to describe learners’ knowledge of subjects, cognitive abilities, subject literacy, learning styles, and emotional states. It is the premise and foundation for providing personalized support services. The general process of learner profiling includes four stages: acquiring learning behavior data, analyzing learning behavior data, extracting user tags, and generating user profiles.

(5) Empowering Adaptive Learning Diagnostics

Current smart education emphasizes a “learner-centered” teaching philosophy even more. However, how to assess and diagnose learners’ true mastery of knowledge and skills remains a global challenge in educational informatization. According to cognitive theory, learning is a process of forming and establishing cognitive structures through learners’ psychological processing and information handling in real problem situations. Learning diagnostics is the process of assessing learners’ cognitive structures through diagnostic testing.

(6) Empowering Personalized Learning Recommendations

Personalized learning is the essential pursuit and value orientation of educational development, as well as the best practice of empowering education with artificial intelligence technologies. However, the exponential growth of learning resources has increasingly exacerbated learners’ “cognitive load” and “learning drift” problems. The contradiction between the vast richness of learning resources and the insufficient supply of personalized learning services is becoming more pronounced. Personalized learning recommendations can suggest appropriate personalized learning resources and learning paths based on learners’ current knowledge states, becoming the key and foundation for achieving precise personalized learning.

(7) Empowering Intelligent Educational Robots

The “Horizon Report” has predicted for several years that teaching robots will become a key technology continuously impacting the field of education. The 2019 “Innovative Teaching Report” published by the Open University in the UK also pointed out that “robot-assisted learning” may become an “innovative teaching method” in the education sector. Educational robots can act as learning assistants or intelligent companions for learners, providing intelligent services such as Q&A, navigation, recommendations, questioning, and social interaction, which play an important role in enhancing learning interest, stimulating motivation, and improving learning outcomes. At the same time, educational robots can assist teachers in teaching monitoring, management, automatic Q&A, and extend teachers’ abilities in expression, knowledge transfer, and communication.

Currently, the robots applied in the education field are mainly dialogue or chatbots. These robots use machine learning and knowledge graph technologies to enable machines to understand human language and respond to learners’ questions or complete specific tasks based on in-depth analysis of human intentions through simple logical reasoning and rule matching processes. For example, dialogue robots are complex systems that integrate language perception, voice recognition, intelligent decision-making, and automatic feedback, involving various artificial intelligence technologies such as natural language processing, knowledge graphs, knowledge reasoning, and reinforcement learning, with knowledge graphs playing a decisive role.

The basic structure of dialogue educational robots is shown in the following image, which mainly includes four parts: natural language processing module, dialogue management module, knowledge graph module, and natural language generation module. The basic processing flow is as follows: the natural language processing module converts the text and voice data input by the learner into the machine’s internal representation. With the assistance of the knowledge graph, based on deep learning technologies, it processes input for entity recognition, entity linking, coreference resolution, and semantic understanding, ultimately parsing it into slot-value pairs; the dialogue management module integrates the learner’s input data and information from the knowledge graph, generating answers to questions through knowledge reasoning, semantic disambiguation, context understanding, and semantic retrieval, which are finally fed back to the current learner by the natural language generation module. In this process, the knowledge graph serves as the brain memory system of the educational robot, storing vast amounts of common knowledge in the field of education, as well as data on learners’ emotional states, interests, preferences, and knowledge skills, enabling the robot to possess memory, thinking, reasoning, and judgment capabilities similar to a human brain.

Basic Structure of Dialogue Educational Robots

Deep learning and knowledge graphs are the latest achievements of the symbolic and connectionist research paradigms of artificial intelligence. Deeply integrating these two technologies with education and teaching to improve the quality of precise teaching and the level of personalized services has become an inevitable requirement for promoting the development of Education Informatization 2.0. This is also an important way for smart education to move from theory to practice. However, relying solely on deep learning technologies for smart education poses challenges such as the black box problem and lack of interpretability. Knowledge graphs precisely bridge this gap by driving the development of smart education with knowledge at its core. As the foundation for the development of smart education from the stage of “perceptual intelligence” to “cognitive intelligence,” knowledge graphs can provide technical support for applications in education such as intelligent processing of educational big data, semantic aggregation of teaching resources, optimization of smart teaching, construction of learner profiling models, adaptive learning diagnostics, personalized learning recommendations, and intelligent educational robots.

Source丨Hexi University

Disclaimer丨This repost is intended for sharing. This account respects original works, and copyrights belong to the original authors. If there are copyright issues, please contact us for deletion.

↑ For more courses, please follow “Duxing Education Research Institute” ↑

Applications of Educational Knowledge Graphs in the Context of AI+

Recommended Readings

Leave a Comment Cancel reply