IRSep 25, 2020

A review of metadata fields associated with podcast RSS feeds

arXiv:2009.12298v24 citations
Originality Synthesis-oriented
AI Analysis

This work highlights data quality issues in podcast metadata that affect developers building recommendation systems, though it is incremental as it documents existing problems rather than proposing new solutions.

The researchers investigated how metadata fields in podcast RSS feeds are actually used by creators, finding that key fields like itunes:type, season number, and category are often not used as expected, which could negatively impact recommendation systems.

Podcasts are traditionally shared through RSS feeds. As well as pointing to the audio files, RSS gives a creator a way of providing metadata about the podcast shows and episodes. We investigate how certain metadata fields associated with podcasts are currently being used and comment on their applicability to recommendations. Specifically, we find that many creators are not using the itunes:type field in the expected fashion, and that using this field for recommendations might not lead to an optimal user experience. We perform similar explorations for the season number and the category associated with a podcast, and also find that the fields aren't being used in the expected fashion. Finally, we examine the notion that a single podcast show is the same as a single RSS feed. This also turns out to not be strictly true in all cases. In short, the metadata associated with many podcasts isn't always reflective of the show and should be used with caution.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes