Simply knowing whether a customer number has subscribed to Zalo is not enough to understand user characteristics and account status, making it difficult to perform effective data analysis.
The original number list lacks username and other information
Unable to understand the age and gender distribution of customers
Banned accounts will reduce the proportion of effective users
Manually reviewing large amounts of Zalo data is expensive
Lack of unified fields to establish customer labeling system
Speculative attributes can easily be mistaken for definite facts
