This study aims to highlight the inconsistency between conventional knowledge push judgements and the price of knowledge push. Also, a three-way decision-based relevant knowledge push algorithm was proposed.
Using a ratio of 80–20%, the experiment randomly splits the data into a training set and a test set. Each video is used as a knowledge unit (structure) in the research, and the category is used as a knowledge attribute. The limit is then determined using the user’s overall rating. To calculate the pertinent information obtained through experiments, the fusion coefficient is needed. The impact of the push model is then examined in comparison to the conventional push model. In the experiment, relevant knowledge is compared using three push models, two push models based on conventional International classification functioning (ICF), and three push models based on traditional ICF. The average push cost accuracy rate, recall rate and coverage rate are metrics used to assess the push effect.
The three-way knowledge push models perform better on average than the other push models in this research in terms of push cost, accuracy rate and recall rate. However, the three-way knowledge push models suggested in this study have a lower coverage rate than the two-way push model. So three-way knowledge push models condense the knowledge push and forfeit a particular coverage rate. As a result, improving knowledge results in higher accuracy rates and lower push costs.
This research has practical ramifications for the quick expansion of knowledge and its hegemonic status in value creation as the main methodology for knowledge services.
To the best of the authors’ knowledge, this is the first theory developed on the three-way decision-making process of knowledge push services to increase organizational effectiveness and efficiency.
