twitter analysis/ comparison with dataset
Hi
The annoying part of this is I used rapid miner when doing a module with the university of liverpool never touched since and now i have forgotten how to do anything, it is not like ridding a bike!
Could someone please give me a few pointers/ refreshers to achieve the following?
I have put together a twitter connector which i am happy with pulling in my keyword tweets
I have an excel which i have added as a dataset that contains a column with the ‘tweet’ and another column with either a 0 or 1
What i need to do is analyse the tweet coming into the connector and score on comparison to previous tweets that have the 1 in my dataset
The next step will be to use a zap which i am happy with (kind of) to go add a row to a google sheet, ideally i would prefer to not use zap though to some how use rapid miner to add a new row to a google sheet or sheet in box that contains simply all the data from the original tweet and my grading value
Then to automate the whole thing so it runs every 5 minutes
Thanks in advance
Chris
Answers
If I understand this correctly, the Zap contains your historical data which you want use to score the incoming Tweets?
If that's the case, you really don't need the Zap Just export that to an Excel sheet and then use that as your training data. Then build a model, score, and append to the historical data set.
That's it in simplictic terms, like the attached screenshot.
To make it run every 5 minutes, install the RapidMiner Server and run it from there.
Thank you so much for your reply
Yes you are correct my training data if you would can come from excel, all i need zap to do is write out the tweet with the score after it has been analysed
I also seam to be missing a trick I do not how to automate a process so it just runs every 5 minutes or any other schedule
Chris
This part I don't understand, "all i need zap to do is write out the tweet with the score after it has been analysed"
Is your Zap writing thre results somewhere else outside of RapidMiner? You could use the Zapier connector to do that.
If not, I'd just take the process and and write all the predictions back to an Excel. Just attach a Write Excel at the end of hte Append operator.