Fashion and Style Trends

An interesting area for data mining is fashion and style trends. Understanding what people want now and how that influences what people might want in future. How tastes and fashion changes over time. It seems fashion is a very unpredictable affair. Often styles are influenced by multiple factors. Especially, trends develop from word of mouth even by noticing what a celebrity has can create a fad. Perhaps, social trends are important as a measure of predicting or even forecasting to some degree what will be on next so people can design and purchase. WGSN, Trendzine, Editd are a few interesting sources for fashion trend forecasting and analysis. There are obviously others that provide trend intelligence. However, as a commercial service a lot of the data is closed for inspection and accessibility with an added premium. It may be an idea to explore fashion as a semantic ontology by way of curating data. Then providing such data in an open data format for much analytical inspection on the web. Social influence graphs and sentiments could also help in assessing the right factors as to where a particular trend is likely to be in near future as well as individual preferences. Fashion trends are especially appropriate to women and there is a huge market that can be tapped. Better fashion sense means people will be more informed about what they should buy now as opposed to later. Also, it helps shops and designers plan their products and sales more appropriately. It also means people are always aware of what is currently trending in market now and potentially in next twelve months. Obviously, the more long-term the forecast is, the less accurate the information gain will appear.

Using big data to forecast the fashion future
Dress to kill how semantic web can help