Discover Hidden Facebook Insights
Ever wondered how insightful your Facebook posts can be? Let me tell you, if you derive insights from Facebook posts and comments you can actually derive your customer needs and pain-points.
As a result, on September 8th 2020, IBM Developer Advocates Fawaz Siddiqi & Faiz Hasan conducted a webinar about how you can derive and visualize insights obtained from data acquired from Facebook posts and comments.
The webinar was divided into two parts, the first part was conducted by Fawaz in which he talked about AI vs. IA, and IA is very essential as it helps us design a system which actually enhances human cognitive skills to actually interpret data quickly and take decisions quickly based upon the data provided. lement
Then he moved on talking about various other IBM tools which would help us attain our goal to analyze Facebook data. Tools such as IBM Watson Studio which is a single platform for all your data science and machine learning needs, Watson Natural Language Understanding which helps us to check for user sentiments based on a particular text, Watson Visual Recognition to classify image data and Cognos Embedded Dashboards to easily visualize our findings from our data preparation and churning processes which take place within our data science pipeline.
After which Faiz took over for the hands-on session where he showed the audience on how we can implement a solution based on Pixiedust and incorporate all the mentioned IBM services in order to gain more insights from our data.
The flow & architecture of the exercise was the following:
- A CSV file exported from Facebook Analytics is added to Object Storage.
- Generated code makes the file accessible as a pandas DataFrame.
- The data is enriched with Watson Natural Language Understanding.
- The data is enriched with Watson Visual Recognition.
- Use a dashboard to visualize the enriched data and uncover hidden insights.
And the outcome was a dashboard based on the various values mentioned in the Facebook posts such as the mentioned entities and various products, the session was very interactive and the participants were very keen to learn more.
Fawaz & Faiz then discussed various other ways in which the same solution can actually be also used for a real-time analysis of data for an enterprise which uses Facebook and other social media platforms to market their products and services.
Interested in doing this by yourself?
Sign up for IBM Cloud here
Find the hands-on code pattern here
You can find the workshop materials (slides) here
