Promoting Music Exploration through Personalized Nudging in a Genre Exploration Recommender

Recommender systems are efficient at predicting users’ current preferences, but how preferences develop over time is still under-explored. In this work, we study the development of musical preferences. Exploring preference consistency between short-term and long-term in data from earlier studies, find that users with higher expertise have more consistent for their top-listened artists tags than those lower expertise, while high also show diverse listening behavior. Users typically chose to explore genres were close effect was stronger expert users. Based on these findings, conducted a user genre exploration investigate (1) whether it possible nudge distant genres, (2) behavior within influenced by default recommendation settings balance personalization representativeness different ways, (3) nudging increases perceived helpfulness recommendations explore. Our results likely select if nudged presented top list, however, less do so. We representative slider, which recommended genre-representative tracks, made set slider personalized level. The position alone did not promote when combined positions effectively Nudging does necessarily lead an increase helpfulness. On one hand, improves making away other reduces due as move recommendations. To improve helpfulness, seems necessary provide balanced trade-off personalization.