Is it possible to predict successful fashion trends?
\nWith the emergence of Big Data, the biggest challenge for data scientists today is no longer accessing enough data. The difficulty is identifying the right data and data granularity to predict trends accurately, especially when what you\u2019re looking at has high volatility. Also thanks to social media and other factors, fashion trends have become incredibly volatile and as such, increasingly demanding for traditional forecasting approaches that are based on historical data.
\nIn the fashion industry, all stages of the production process – from the acquisition of raw materials to the sale of the finished product – require careful planning and forecasting, because they are influenced by endogenous trends (popularity, prices, promotions and competition) and exogenous trends (seasonality and weather).<\/p>\n
Evo, a start-up based in I3P and London that uses AI to help companies create more effective supply chain and pricing strategies, proposes a challenge to help companies with resource planning by more accurately identifying emerging trends in purchasing behaviour, especially when it comes to style.
\nThe primary goals will be to:<\/p>\n
This activity will be recognized within your educational plan with 8 CFU credits.<\/p>\n
Check your institutional email, look for \u2018Challenge@PoliTo\u2019 and read the regulation carefully.<\/p>\n