A knowledge graph-integrated recommendation method for college student career planning

Low recommendation accuracy: Current college student career planning recommendation methods often have low accuracy when handling complex and diverse student interests.

Low recommendation accuracy: Current college student career planning recommendation methods often have low accuracy when handling complex and diverse student interests.

Knowledge graph-integrated recommendation model: The paper proposes a career planning recommendation model that integrates a constructed career planning knowledge graph with latent user and item representations.

GCN neighbor aggregation: The approach uses graph triplet information to form neighborhoods and applies a Graph Convolutional Network to aggregate neighbor vectors into enriched student and career representations.

KG-GMF representation fusion: The model fuses graph-based representations with generalized matrix factorization representations and scores recommendations through a fully connected layer.

Knowledge graph-integrated recommendation model: The paper proposes a career planning recommendation model that integrates a constructed career planning knowledge graph with latent user and item representations.

GCN neighbor aggregation: The approach uses graph triplet information to form neighborhoods and applies a Graph Convolutional Network to aggregate neighbor vectors into enriched student and career representations.

Higher recommendation accuracy:

Higher recommendation accuracy: