Research And Implementation Of Correlation Strategy Between Video Resources And Knowledge Points Based On Collective Wisdom Annotation

In the background of big data in education,the number of digital learning resources has exploded.Learmers are easily lost when they browse resources.Or they waste a lot of time to find the resources they want.Maybe they cannot find them finally.Among many types of resources,video resources integrate information such as text,images,animations,and sounds.They transmit information from multiple dimensions.Learners can concentrate on watching and listening while browsing video resources.At present,for the above problems and the characteristics of video resources,this paper proposes a strategy for constructing the relationship between video resources and knowledge points based on public intelligence annotation.Designing and implementing the module to dig out the potential knowledge points in the entertainment video resources,and users can learn knowledge by watching the entertainment video resources.The main work of this paper includes:(1)Based on semantic analysis,the relationship between resources and knowledge points is constructed.It mainly discusses from two aspects.The relationship is calculated based on STT and semantic analysis and the relationship is calculated based on resource similarity.(2)Based on public intelligence annotation,the relationship between resources and knowledge points is constructed.It mainly discusses from two aspects:description information and related topic words.For the description information,obtaining the topic keywords through the LDA theme model,calculating the similarity between each topic and the knowledge points,and then constructing the relationship.For the association keyword information,the user domain confidence is integrated to realize the association between the video resource and the knowledge point.(3)The self-growth mechanism of the knowledge system is mainly carried out from two aspects.On the one hand,the self-growth of the knowledge system is based on the semantic analysis of the description information.On the other hand,the self-growth of the knowledge system is based on the knowledge points added by the user in the knowledge system and the keywords added by the user when labeling the relationship.Finally,the experiments verify the validity and reliability of the association strategy.And relying on the resource integration platform,the public intelligence labeling module was designed and implemented.Functional testing verifies the effectiveness of the module.