Papers
Topics
Authors
Recent
Search
2000 character limit reached

Style in the Age of Instagram: Predicting Success within the Fashion Industry using Social Media

Published 18 Aug 2015 in cs.CY, cs.SI, and physics.soc-ph | (1508.04185v1)

Abstract: Fashion is a multi-billion dollar industry with social and economic implications worldwide. To gain popularity, brands want to be represented by the top popular models. As new faces are selected using stringent (and often criticized) aesthetic criteria, \emph{a priori} predictions are made difficult by information cascades and other fundamental trend-setting mechanisms. However, the increasing usage of social media within and without the industry may be affecting this traditional system. We therefore seek to understand the ingredients of success of fashion models in the age of Instagram. Combining data from a comprehensive online fashion database and the popular mobile image-sharing platform, we apply a machine learning framework to predict the tenure of a cohort of new faces for the 2015 Spring\,/\,Summer season throughout the subsequent 2015-16 Fall\,/\,Winter season. Our framework successfully predicts most of the new popular models who appeared in 2015. In particular, we find that a strong social media presence may be more important than being under contract with a top agency, or than the aesthetic standards sought after by the industry.

Citations (60)

Summary

Predicting Success in Fashion Modeling through Social Media Activity

The research paper titled "Style in the Age of Instagram: Predicting Success within the Fashion Industry using Social Media" presents an empirical study focused on understanding the dynamics of success for fashion models in a modern context intertwined with social media platforms, particularly Instagram. The study is driven by two main research questions aiming to uncover whether measurable physical and professional characteristics can forecast casting success in fashion modeling and if the integration of social media indicators enhances the predictability of such success.

Methodology and Data

The methodology employed in the paper combines traditional physical attributes and professional indicators with novel social media metrics to predict the tenure and popularity of fashion models. Data for this research was gathered from the Fashion Model Directory (FMD) and Instagram. The FMD data included physical attributes such as height, hip size, dress size, waist size, and shoe size, alongside professional details like agency representation. Instagram data focused on social media presence, namely the number of posts, likes, comments, and sentiment analysis of comments, over specific timeframes aligned with fashion show schedules.

The analytical approach involved a Poisson regression framework to estimate the impact of these variables on the number of runways walked by models in major fashion weeks. Machine learning techniques, including Decision Trees (DT), Random Forest (RF), and AdaBoost (AB), were deployed to perform predictive classification of model success during subsequent fashion seasons.

Key Findings

The study revealed several critical insights into the determinants of success in fashion modeling:

  • Physical Attributes: Taller models tend to have higher chances of success, whereas larger dress, hip, and shoe sizes negatively impact runway walk counts. This aligns with the industry's aesthetic standards favoring slender physiques.
  • Agency Representation: Being represented by a top agency significantly boosts a model's runway opportunities, underscoring the role of established agencies in setting trends and elevating models' careers.
  • Social Media Influence: Surprisingly, social media metrics such as the number of posts and sentiment derived from comments provide meaningful predictive power, often matching the influence of professional representation. Higher activity levels on platforms like Instagram can amplify a model's visibility and, consequently, their likelihood of being featured in prestigious shows.
  • Machine Learning Predictions: The Random Forest model demonstrated the highest accuracy in predicting model success, achieving an AUROC of over 81% in real-world prediction tasks for future fashion seasons. This confirms the feasibility of using combined datasets for reliable forecasting in cultural industries driven by popularity.

Implications and Future Directions

The findings emphasize the evolving nature of success in industries heavily influenced by cultural trends and collective social perceptions. Social media has surfaced as a valuable tool not only for marketing within the fashion industry but also as a predictive instrument that reflects real-world events and success potentials.

The paper suggests several pathways for future research, including enriching the measures of success beyond runway counts to include other dimensions like magazine features and social engagements. Moreover, the study showcases potential for extending this methodology to other cultural markets driven by prestige, such as art and music, or even domains like scientific production where social influences and word of mouth play significant roles.

In conclusion, as computer-mediated communications continue to transform cultural production landscapes, methodologies integrating social signals offer promising avenues for understanding and forecasting success dynamics in the fashion industry and beyond.

Paper to Video (Beta)

No one has generated a video about this paper yet.

Whiteboard

No one has generated a whiteboard explanation for this paper yet.

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Continue Learning

We haven't generated follow-up questions for this paper yet.

Collections

Sign up for free to add this paper to one or more collections.