Sentiment Analysis for Demand Forecasting
In the article Demand Forecasting with exogenous data, we introduced the Sentiment Analysis. Basically, it studies sentences coming from the Internet (think about posts on Twitter, Facebook, Google Plus ecc…) in order to understand people sentiments.
Why is Sentiment Analysis so important?
In order to answer the first question, let us think about an example. Suppose you are a fashion company and you want to know the next month sales of your leather bag. If Twitter reveals that there are an increasing number of positive posts talking about your bag, maybe you will have more boosted sales than you expect.
How can Sentiment Analysis data be used for demand forecasting?
First of all, you have to catch all the posts talking about a topic of your interest. For example, you can obtain tweets talking about your bag by using hashtags. Now you are able to extract sentiments from sentences (happiness, sadness etc).
Then, you have to count the number of happy/sad posts talking about your leather bag. Those numbers can be used with the historical data in order to obtain a proper demand forecasting.
In conclusion, Sentiment Analysis is very important for companies for having a demand forecasting considering global trends and sentiments. However, the correct usage of those data is the main challenging task, but it makes the difference between a bad and a good demand forecasting, and Intuendi knows it.