This project involves web scraping Ericsson's Facebook page (https://www.facebook.com/ericsson/?brand_redir=290266184368865) using the facebook_scraper module in Python. The collected data is then subjected to sentiment analysis using the vaderSentiment module to classify the posts as positive, negative, or neutral. The scraped data is then stored in an Excel file for further analysis.
- Introduction
- Data Collection
- Sentiment Analysis
- Data Analysis with Tableau
- Visualizations and Insights
- Business Implications
Social media plays a vital role in understanding customer sentiment and engagement. In this project, we analyze Ericsson's Facebook page to gain insights into the performance of their posts and understand how customers are interacting with the content.
Using the facebook_scraper module, we scraped Ericsson's Facebook page to retrieve information such as post text, number of likes, comments, and shares, and the post timestamp. The data collection process allowed us to obtain a comprehensive dataset for further analysis.
To better understand the sentiment of the posts, we performed sentiment analysis using the vaderSentiment module. This allowed us to classify the posts into positive, negative, or neutral categories based on their content.
After data collection and sentiment analysis, we stored the scraped data in an Excel file for easy accessibility. Subsequently, we utilized Tableau for in-depth data analysis and visualization. Tableau's powerful features allowed us to create interactive and insightful visualizations.
The visualizations derived from Tableau provided us with valuable insights, including:
-
Maximum number of posts in February 2023 (25 posts) indicating high activity during that period.
-
Average number of comments, likes, and shares per month, providing an understanding of post engagement over time.
-
Identification of the top 10 most engaging posts based on the number of likes, enabling us to understand which content resonated most with the audience.
-
Sentiment analysis showed that 172 posts were classified as positive, 14 as neutral, and only 1 as negative, indicating an overall positive sentiment towards Ericsson's Facebook content.
The visualizations and insights derived from the data analysis can have several implications for Ericsson's social media strategy:
Understanding peak posting periods can help optimize content scheduling to reach a larger audience. Analyzing post engagement metrics can guide the creation of content that encourages more likes, shares, and comments. Identifying the most engaging posts can provide valuable insights into the type of content that resonates best with the audience. Monitoring sentiment analysis helps the company gauge customer feedback and address any potential negative sentiment promptly. By leveraging these visualizations and insights, Ericsson can enhance its social media presence, engage with its audience more effectively, and make informed decisions to improve its overall online reputation and brand image.
Note: The data used in this project is collected from Ericsson's public Facebook page, and all analyses are based on publicly available information.
Contributions to the Ericsson's Facebook Analysis are welcome. If you have any suggestions, improvements, or feature additions, please feel free to submit a pull request.