WhatsApp Group Chat
After completing the Google Data Analyst Professional Certificate on Coursera, I decided to advance my skills in Business Intelligence, hence chose to dive deeper into Power BI, although Tableau was used in the Google Certification Programme. I then took courses, Power BI Essential Training and Learning Power BI Desktop, all by Gini Von Courter on LinkedIn Learning and wanted to put my learning into practice. So here is my first Project in Power BI. Please enjoy. Please continue to read how the project was done or you can just visit Project Here to interact with it.
Objective: Identify trends in WhatsApp Group Chat.
Task: Provide a Dashboard of Group Chat analyses.
Dataset: Dataset was obtained from the Group via the Export command.
Data Processing
Data Cleaning: Group chat was exported in csv file and further cleaned in Microsoft Excel. The chats had date, time, contact and message all in a single row per message. Total of 28,000 rows had to be cleaned. These rows were then converted to columns using Excel’s Text to Column command. After converting the texts to columns, I realised some rows had no date, time and contacts, so I deleted those empty rows.
Data De-Anonymization: A new sheet for contacts was created for Contacts. Duplicates were removed from the Contacts in order to have a single distinct contact. User IDs were further created for each user. The IDs were randomly created using the randbetween function. These IDs were copied and pasted as values. Duplicates in IDs were highlighted by the help of Conditional Formatting. This made it easier to replace IDs that were duplicated.
Date & Time: Further analysis of date and time were done in Power BI since it was much flexible. The data was for the same year therefore, there was no need to extract the year. However, Month Name and Day Name were extracted. Hours were also extracted from the Time.
Insights
The group was made up of 120 Members. Over 28,000 messages, excluding media, were shared on the platform between January and August. Total Media uploads to the group were over 2,000 media files. Members of the group also showed interest in sharing external links to the group.
While messages peaked on Tuesdays, the month of March was the month with most messages. It is fascinating to know that March recorded the largest message because members’ discussions were mostly about Ghana’s Independence Day, which takes place on 6th March. The discussion was centered on the usefulness or otherwise, of the independence day celebration.
It was revealed that top 10 members made the group active, without whom the group would have been dead. Majority of the group members do not chat in the group. This needs further investigation so that the reasons for which members do not chat in the group would be known and solved, so that everyone on the group can chat freely and be happy while on the group.
Conclusion
This project pushed me to my limits as a newbie in Power BI, and I hope to get better at it as time goes on. The project has exposed me to tons of functionalities that Power BI offers and has expanded my skillset. A big thank you to my mentor, Alex Freiberg, and all other data community members who keep sharing their experiences.
I hope you enjoyed this project as much as I enjoyed building it.
Please interact with the project here and kindly let me know what you think.