The human mind is an unparalleled pattern recognition machine. Wordclouds are a visualisation technique which allow the human eye to pick out patterns with a little help from statistics. For this first post the Human mind will be the machine.
What can you pick out of the clouds?
Post Titles in July
Blockchain
When I look at Steemit, I see a rich public datasource. An immutable public Blockchain that is waiting to be explored.
Data in the 21st Century
All facets of life are present here; cultures, languages, religions, occupations. One of the greatest aspects of Steemit is the fact that it is on a public blockchain. The data is available for anyone to view and immutable. It cannot be changed, censored or hidden. As Steemit grows it can become difficult to keep track of the authors you really like. We get flooded with data. I am now following over 900 people so my feed is usually full of posts, too many for me to reasonable read and respond to in a day.
Post Body Text in July
Machine Learning
The machines are rapidly catching up with the human brain and in this series of posts I will be using advanced machine learning techniques to find patterns in the Steemit Blockchain, establishing relationships and connections. It will also help me to find posts and authors to follow.
Thanks to @furion and steemdata.com the blockchain data is publicly accessible and using some cutting edge machine learning techniques I will explore the blockchain over the coming weeks, looking for interesting patterns in the data.
Do you want to know
- Who is the most positive author?
- What combination of tags correlate to the highest payouts?
- What is the best time of day to post, for comments, payout rewards, votes?
- What topics draw the largest comments, payouts and votes?
Machine Learning Challenges
Anyone interested in data science will have heard of https://www.kaggle.com/.
Wouldn't it be interesting to submit Steemit Blockchain data to this site for a competition with a bounty?
If anyone has any good ideas of a problem, and wishing to support a bounty I would love to hear your suggestions.
Recent Related Posts
- Followmenomics "The Follow Me Index"
- Who does all the Chatting?
- Steem Payout Patterns
- Steemit Payouts and Author Reputation
- Post Payouts Since Hf19 Update
I am planning an update to this last post in the next few days.
Thank you for reading. I write on Steemit about Blockchain, Cryptocurrency, Travel and lots of random topics.
That's a really nice idea. Maybe some algorithm to predict how successfully a post is, or maybe a recommending system for posts/authors.
I used kaggle a long long time ago when I was learning about machine learning and steem blockchain could be a goldmine for any machine learning enthusiast.
Kaggle has a section where you can publish datasets.
Downvoting a post can decrease pending rewards and make it less visible. Common reasons:
Submit
steem on
Downvoting a post can decrease pending rewards and make it less visible. Common reasons:
Submit
kaggle looks very cool. Will join very soon. Presently i am researching about data science.
Downvoting a post can decrease pending rewards and make it less visible. Common reasons:
Submit
An interesting topic with a bounty on it, would get great interest from that community and would be a good way to attract people to Steemit.
Downvoting a post can decrease pending rewards and make it less visible. Common reasons:
Submit
Great post! ;)
Downvoting a post can decrease pending rewards and make it less visible. Common reasons:
Submit
steemit, love, bitcoin,
Downvoting a post can decrease pending rewards and make it less visible. Common reasons:
Submit
We sure do love talking about steemit.
Downvoting a post can decrease pending rewards and make it less visible. Common reasons:
Submit
No doubt!
Downvoting a post can decrease pending rewards and make it less visible. Common reasons:
Submit
Awesome post! We see wordclouds all the time, but seldom hear about the technology behind these awesome visualization techniques. I've also dabbled a bit with wordclouds -- if anyone wants to learn more about them, check out my post here. I hope I see more data visualization posts going around!
Downvoting a post can decrease pending rewards and make it less visible. Common reasons:
Submit
Good post @eroche
Follow me @Julfan
Downvoting a post can decrease pending rewards and make it less visible. Common reasons:
Submit
One thing that pops out from the title word cloud is that positive words vastly outnumber negative words.
Downvoting a post can decrease pending rewards and make it less visible. Common reasons:
Submit
My first proper machine learning exercise is going to be sentiment analysis. I am going to try to find all the most positive authors.
Downvoting a post can decrease pending rewards and make it less visible. Common reasons:
Submit
kaggle rocks. So does data science and machine learning. What did you use for this? Did you use Power BI word could visualization?
Downvoting a post can decrease pending rewards and make it less visible. Common reasons:
Submit
I haven't yet, I will give it a try on this data and see what kind of results I get.
I wonder can it handle non latin text?
Downvoting a post can decrease pending rewards and make it less visible. Common reasons:
Submit
i don't think it can handle non latin text
You can run sentiment analysis too using power bi and microsoft cognitive services...I have a post on my website on how to do it
Downvoting a post can decrease pending rewards and make it less visible. Common reasons:
Submit
interesting
Downvoting a post can decrease pending rewards and make it less visible. Common reasons:
Submit