Extreme Multi-tagging via Deep Learning and Explanation techniques

What is it about?

Extreme Multi-tagging consists in indexing multimedia content with rich metadata, in order to improve the quality of the search engines’ results and boost the recommender systems' performances. In this context, we devised a Deep Learning based solution for semi-automatic multi-label classification, further integrating post-hoc explanation techniques.

Featured Image

Why is it important?

The adoption of tools able to automatically label multimedia content with informative tags represents an important task for all the companies offering streaming entertainment services. However, domain experts generally perform the tagging process manually, making it time-consuming and error-prone. Our Deep Learning solution copes with the issues described above, while the model explanation methods are exploited to assist the operator in the labeling process by facilitating an understanding of the model predictions.