Case study from #msmbc10 – Sentiment Analysis and Qualitative Research

  Archives Categories

Case study from #msmbc10 – Sentiment Analysis and Qualitative Research


The debate for sentiment analysis has led to the conclusion for many like Nathan Gilliatt that sentiment is something that must be analyzed by a person. At the Monitoring Bootcamp in London we included one case study from one of our clients that was able to benefit from human analysis.

There has to be someone behind it that can filter through your search returns to analyze the articles for sentiment both at a global and at a post level.

We’d like to share with you, then, one of the case studies that Michelle shared with the crowd at #msmbc10 to get your feedback and opinions on human-analyzed sentiment.

The case : one product, one spokesmodel, 2 countries, 2 languages

-For one client, for example, they wanted to evaluate the relationship between a chosen spokesmodel and a new product launch
-Comments were marked as « negative » if they mentioned both the spokesmodel and the product in a negative way
-Positive meant the oppostie
-And neutral was mixed
-But that was just the beginning
-We then looked at differences between different countries where it was launched to give more understanding to these classifications
Different sentiment reactions for each country
The analysis : underlying cultural differences caused differences in opinion
The takeaway : use social media qualitative research to gain insights

Back to Blog

Leave a Reply