Looks like I'm on a roll! Please see my Rapidminer 5.0 Video Tutorial #2. In this video we begin the process of recreating my original written NMT YALE/Rapidminer tutorials into version 5.0 and into a video. This video shows how to import training and prediction data, add a classification learner, and apply the model, and get the results.
Video download link (HQ): Rapidminer 5.0 Video Tutorial #2
The data files you will need to follow along are Excel spreadsheets below:
Training data set: gold_final_input
Prediction data set: ga-gold
Thanks for this. I can't get enough of your RM help!
CBedo: Thanks. I hope you put these tutorials to good use!
Fantastic Videos Tom!!!
I have using Rapidminer for a couple of months to make a model for predicting Forex prices and your videos have been a tremendously helpful resource. I am looking forward to new video series.
Thanks,
Cleo
@Cleo: I've had great success using RM to model Forex trends. Read up on my $100 Forex Experiment, that was a trend system developed with RM.
Kudos Tom,
Thanks for sharing; continued stamina and success to you!
Mike
Thanks. Great little video showing clear steps and reproducible results.
@Mark: That’s been the intent of my videos all along!
hi Tom,
how do you sync your data? For example the DJ and the GDax? There are day at which only one of these is traded.
best regards,
StevenD
@StevenD: I use TraderXL to snyc up the data based on dates. If the S&P500 trades, say on 5/5/10 and the GDAXI does not, the GDAXI entry for 5/5 will show 5/4′s closing number.
Hey Thomas, what assets do the CD and EM tickers stand for?
I wish all tutorials were so clear and dense with information! Thanks for your efforts to share and inform – you have helped me.
Great Video. Thanks a lot.
But I have had a little problem when I’m trying to run this process but I got a warning that prevent the process to complete the generation of the result.
The warning is: Defaul Model: Cannot use method ‘median’ for nominal labels. Change to ‘mode’!
After this first warning, I got a list of some others warning, one of which is: The number of nominal values is not the same for training and application for attribute ‘^TNX’, training: 257, application 7.
For sure I am in wrong somewhere, Could you help to solve it?
Thanks again, pb
Thanks pb_. It appears that somewhere in your data there’s missing nomimal data. RM tries to use a median value for the nominal data but it can’t because there is no median value between “UP” or “DOWN.” Try reviewing the input data again for any obvious errors, I suspect the second warnings would go away then.
Thanks a lot man! I am on my second year Masters program in IT and my project or thesis is “Open Source Data Mining applicability in Higher Education: Investigating Rapid Miner” Watching your video has been very informative. Will surely cite your website and you in my thesis.
@Mesfin: Thanks
Wow, you’re really good. Great step by step approach.
@tony: Thanks.
Each of your videos has a lot of excellent material in them, hopefully you will find it worthwhile to make several more as time passes. There are few videos on the web that can come close to yours for clarity.
Thanks for the work you have put into these
DaveM: Thanks for your kind words, I’m hoping to record some new ones in the Spring. I seem to be on a one year cycle for this. =)
Hi, thanks for videos.
I am looking for using Rapid Mine with forex GBPUSD daily/weelly/H4 prediction.
How can I make this with Rapid Mine ?
Thanks in advance!
AlbertoFX you ask a simple question that has a very complicated answer. I suggest you watch my time series tutorials, that’ll give you a starting point.
I would like to ask this Tutorial Part 1 and 2, using validation model and using just a simple model to prediction. When I ask, add some training data (“UP”) into testing data, the result still all “Down”. And I copy all the training data into testing data, the result are still predict to “Down”. Actually, I follow the steps of tutorials. Can anyone explain why?
Sunny, the example I show in the tutorial isn’t a very good model to use for prediction. It’s accuracy is like 50/50 which is a coin toss. There is no magic formula to wealth here. ;)
Hi Neuralmarkettrends,
So would you mind providing another example that can show a good model to use for prediction.
Sunny
Hi Neuralmarkettrends,
So what should I change the model so that some results will be “Up”? I am a beginner of this. Thank you very much for your help.
Sunny
Well you can try replacing the model operator to something like Support Vector Machine or you could add more input variables to help the model learn the classification better.