KeNet:Knowledge-enhanced Doc-Label Attention Network for Multi-label text classification

arXiv:2403.01767v1 Announce Type: new
Abstract: Multi-Label Text Classification (MLTC) is a fundamental task in the field of Natural Language Processing (NLP) that involves the assignment of multiple labels to a given text. MLTC has gained significant importance and has been widely applied in various domains such as topic recognition, recommendation systems, sentiment analysis, and information retrieval. However, traditional machine learning and Deep neural network have not yet addressed certain issues, such as the fact that some documents are brief but …

abstract arxiv attention classification cs.cl domains importance knowledge labels language language processing multiple natural natural language natural language processing network nlp processing recognition recommendation recommendation systems sentiment systems text text classification type