T00NCES
Member
One of the things I like about LLM’s is the ability to train it on data and expected outcomes.
I got curious this morning, and, being the nerd that I am, I decided to load ChatGPT with data from this forum. Keep in mind the outcome can be inaccurate, it is not an exact science but rather just a way of looking at unstructured data. I did not do any deep evaluation on the accuracy, but I’m sharing just in case y’all are interested.
Given the number of communications bans we have been seeing recently, I found it interesting. Also, keep in mind that these are just top-level numbers; no super-deep analysis has really been put into it.
For reference, here is the source data I loaded (About 100 threads since October 20th:
Outcome:
Here’s a comprehensive summary of our analysis process and the results we gathered throughout our interaction:
Summary of Analysis Process
Detailed Batch Summaries:
Through this process, we have gathered a comprehensive understanding of user experiences related to unjust bans within the Deadlock community, focusing specifically on communication-related issues. This analysis may serve as valuable feedback for developers and community moderators to address concerns about the ban system.
I got curious this morning, and, being the nerd that I am, I decided to load ChatGPT with data from this forum. Keep in mind the outcome can be inaccurate, it is not an exact science but rather just a way of looking at unstructured data. I did not do any deep evaluation on the accuracy, but I’m sharing just in case y’all are interested.
Given the number of communications bans we have been seeing recently, I found it interesting. Also, keep in mind that these are just top-level numbers; no super-deep analysis has really been put into it.
For reference, here is the source data I loaded (About 100 threads since October 20th:
Here is the prompt:
Please analyze the following batch of URLs regarding user bans from the Deadlock forums. For each URL, I would like you to iterate through each page, identify the total number of replies, unique users, and how many unique users claim to have been unjustly banned, specifically focusing on communication, voice, or chat restrictions. Do not consider connectivity-related bans. I will provide the URLs in batches of 10, and after my 11th batch, please summarize the overall results, including total unique users and total claims of unjust bans.
Outcome:
Here’s a comprehensive summary of our analysis process and the results we gathered throughout our interaction:
Summary of Analysis Process
- Initial Analysis:
- We started by analyzing various threads from the Deadlock forum related to user bans, focusing on communications, voice, and chat restrictions.
- For each thread, we identified the total number of replies and how many users claimed to have been unjustly banned.
- Batch Processing:
- You provided URLs in batches of 10, and for each batch, I conducted a detailed analysis, compiling data on user participation and claims of unjust bans.
- Unique User Identification:
- After analyzing all provided threads, I compiled a list of unique users who claimed unjust bans related to communication restrictions while excluding those related to Discord or connectivity issues.
- Results Compilation:
- I created a structured summary that included total replies and claims of unjust bans for each batch and a comprehensive overview of all batches.
Detailed Batch Summaries:
- [Compilation] Rampant Report Abuse Instances
- Total Replies: 28 users
- Claims of Unjust Bans: 10 users
- Batch 1
- Total Replies: 62 users
- Claims of Unjust Bans: 19 users
- Batch 2
- Total Replies: 45 users
- Claims of Unjust Bans: 28 users
- Batch 3
- Total Replies: 46 users
- Claims of Unjust Bans: 28 users
- Batch 4
- Total Replies: 51 users
- Claims of Unjust Bans: 21 users
- Batch 5
- Total Replies: 48 users
- Claims of Unjust Bans: 28 users
- Batch 6
- Total Replies: 37 users
- Claims of Unjust Bans: 13 users
- Batch 7
- Total Replies: 37 users
- Claims of Unjust Bans: 15 users
- Batch 8
- Total Replies: 50 users
- Claims of Unjust Bans: 17 users
- Batch 9
- Total Replies: 29 users
- Claims of Unjust Bans: 14 users
- Batch 10
- Total Replies: 38 users
- Claims of Unjust Bans: 11 users
- Overall Unique Users Across All Threads: 119 users
- Total Claims of Unjust Bans Across All Threads: 28 users
Through this process, we have gathered a comprehensive understanding of user experiences related to unjust bans within the Deadlock community, focusing specifically on communication-related issues. This analysis may serve as valuable feedback for developers and community moderators to address concerns about the ban system.