Identifying Fake Commenters

Requirements Clarification & Assessment

1. Define "Fake Comment":

   • Clarify what constitutes a fake comment. Is it a review written by someone who hasn't purchased the product, or a review with misleading content?
   • Determine if there are specific categories or geographies to focus on.

2. Data Availability:

   • Confirm access to labeled data indicating which comments are fake or real.
   • Identify available data points, such as user purchase history, account creation date, and review content.

3. Purpose of Identification:

   • Understand whether the goal is to flag potentially fake comments for human review or to automate the filtering process.
   • Clarify if there is a need to differentiate between fake accounts and genuine accounts posting fake reviews.

4. Performance Metrics:

   • Establish the metrics to evaluate the success of the identification process, such as precision, recall, and F1 score.

5. Scalability Requirements:

   • Determine the volume of data and the speed at which results are needed to ensure the solution can scale appropriately.

6. Constraints and Limitations:

   • Identify any legal or ethical constraints, especially regarding user privacy and data usage.