Ai bot detection -- 2

Job Description:

I have a Python Flask web application that allows the user to input a Twitter profile using the profile link or a tweet link. It then uses a trained machine learning model to determine a likelihood of that account being a bot account or not.

I would like someone to use this app and determine its effectiveness in actually identifying bot accounts, test its performance, check for any trends that occur, do some things get flagged more than others, are some checks invalid or inaccurate, how to improve the algorithm and any other things you deem good to note.

Upon receiving the source code, you would do things like:

Thoroughly test the application using a diverse set of Twitter profiles and tweets to evaluate its performance and identify any trends or biases.

Assess the accuracy of the algorithm by comparing its predictions with known bot and genuine accounts, as well as analyzing any false positives or negatives.

Investigate the validity and relevance of the features used in the machine learning model and suggest improvements or additional features that can enhance its accuracy.

Provide recommendations on optimizing the algorithm and overall user experience of the web application.

Deliver a concise and insightful report outlining my findings, including performance metrics, trends, potential issues, and suggested improvements.

If this sounds good, please get in touch. I can provide the entire source code so you can run the web app locally.

Taidot: Python, Tietojärjestelmäarkkitehtuuri, Ohjelmistojen testaus, Machine Learning (ML), Keinoäly

Projektin tunnus: #36261040