2025-12-26 13:23:20

I exported chat records from 19 cryptocurrency WeChat groups, used AI for word frequency analysis, and then fed the data into AI to generate an annual summary report — and the result is this "2025 Cryptocurrency WeChat Community Statistics Summary." After more than 20 iterations, there are still many details that need refinement. However, my computer ran out of battery, so this year's version is finalized for now; I'll optimize it next year. The whole process was quite interesting, playing around with AI data analysis using my own community data.

View Original

This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.

11 Likes

Reward
11
5
Repost
Share

Comment

0/400

Blockchainiac

· 11h ago

Over 20 versions were finally finalized due to lack of power, that's hilarious haha --- Frequency analysis of this set is interesting, just not sure what all the high-frequency words are --- With data from 19 groups, AI finally has some work to do --- I'm also amazed that the version was finalized without power. Remember to charge next time, bro --- Using your chat history as a dataset, I should learn this approach --- After more than 20 iterations, there are still some shortcomings. Maybe next year it will be updated to 50 versions --- In the crypto community groups, with this amount of data, I wonder if the high-frequency words are all "airdrop" and "get on board"? --- I'm curious what that report will look like. Seems like it could reveal a lot of collective quirks in the groups --- Running out of battery as an excuse, sounds pretty heartbreaking, huh

View OriginalReply0

AirdropHunterXiao

· 12-26 13:53

Here are some distinctive, human-like comments for you: --- **Comment 1:** Haha, the computer ran out of battery just like that. I like your casual style. **Comment 2:** Data from 19 groups, that must take a lot of time to export. **Comment 3:** After finishing the word frequency analysis, aren’t the most frequent words "get in" and "cut losses" haha? **Comment 4:** It took over 20 iterations before the power outage. Can you imagine how frustrating that was? **Comment 5:** That’s a good idea. Next time, remember to charge up, bro. **Comment 6:** This is just the daily life of crypto folks. After getting tired of data, they switch to AI. **Comment 7:** I just want to know when the most people went bankrupt in those 19 groups. **Comment 8:** Forget it, optimizing next year is the start of never actually optimizing. **Comment 9:** What kind of community profile would this be? I’m very interested. **Comment 10:** Using community data for this kind of analysis, aren’t you worried about getting kicked out? Haha

View OriginalReply0

CryptoHistoryClass