In many real-world data applications, often we encounter scenarios where each data point may belong to multiple classes. A multilabel classifier is trained to predict the K most likely classes among N possible classes. The article focuses on solving multi-label text classification problems using the Hierarchical Attention Network.
Read more