Search Results

Original Uploader

Simon Fischer

Image Mining with RapidMiner (v1)

Created: 28/04/10 @ 11:00:37 | Last updated: 28/04/10 @ 11:01:04

License: Creative Commons Attribution-No Derivative Works 3.0 Unported License

This is an image mining process using the image mining Web service provided by NHRF within e-Lico. It first uploads a set of images found in a directory, then preprocesses the images and visualizes the result. Furthermore, references to the uploaded images are stored in the local RapidMiner repository so they can later be used for further processing without uploading images a second time.

Rating: 0.0 / 5 (0 ratings) | Versions: 1 | Reviews: 0 | Comments: 1 | Citations: 0

Viewed: 183 times | Downloaded: 90 times

Original Uploader

Sebastian land

Using Remember / Recall for "tunneling" results (v1)

Created: 29/04/10 @ 16:07:55

License: Creative Commons Attribution-No Derivative Works 3.0 Unported License

This process shows how Remeber and Recall operators can be used for passing results from one position to another position in the process, when it's impossible to make a direct connection. This process introduces another advanced RapidMiner technique: The macro handling. We have used the predefined macro a, accessed by %{a}, that gives the apply count of the operator. So we are remembering each application of the models that are generated in the learning subprocess of the Split validation. Af...

Rating: 0.0 / 5 (0 ratings) | Versions: 1 | Reviews: 0 | Comments: 0 | Citations: 0

Viewed: 32 times | Downloaded: 10 times

Original Uploader

Simon Fischer

Stacking (v1)

Created: 29/04/10 @ 16:53:35

License: Creative Commons Attribution-No Derivative Works 3.0 Unported License

RapidMiner supports Meta Learning by embedding one or several basic learners as children into a parent meta learning operator. Here, we use a three base learners inside the stacking operator: decision tree induction, linear regression, and a nearest neighbours classifier. Finally, a Naive Bayes learner is used as a stacking learner which uses the predictions of the preceeding three learners to make a combined prediction.

Rating: 0.0 / 5 (0 ratings) | Versions: 1 | Reviews: 0 | Comments: 0 | Citations: 0

Viewed: 69 times | Downloaded: 30 times

This Workflow has no tags!

Original Uploader

Simon Fischer

Transaction Analysis Demo from RM 5 Intro Day (v1)

Created: 30/04/10 @ 08:19:39 | Last updated: 05/05/10 @ 09:58:51

License: Creative Commons Attribution-No Derivative Works 3.0 Unported License

This is the demo process presented at the RapidMiner 5 Intro Day. It combines customer segmentation with direct mailing. It loads some transaction data, aggregates and pivotes the data so it can be used by a clustering to perform a customer segmentation. Then, additional data is joined with the clustered data. First, response/no-response data is joined, and them some additional information about the users is added. Finally, customers are classified into response/no-response classes. The dat...

Rating: 0.0 / 5 (0 ratings) | Versions: 1 | Reviews: 0 | Comments: 1 | Citations: 0

Viewed: 123 times | Downloaded: 66 times

This Workflow has no tags!

Original Uploader

Ninoaf

test (v1)

Created: 10/05/10 @ 17:04:57

License: Creative Commons Attribution-No Derivative Works 3.0 Unported License

<table><tr><td><p>This process starts with loading the data. After finishing the input operator a typical learning step is performed. Here, an implementation of a decision tree learner is used which also can handle numerical values (similar to the well known C4.5 algorithm).</p></td><td><icon>groups/24/learner</icon></td></tr></table><p>Each operator may demand some input and delivers some output. These in- and outp...

Rating: 0.0 / 5 (0 ratings) | Versions: 1 | Reviews: 0 | Comments: 0 | Citations: 0

Viewed: 18 times | Downloaded: 1 time

This Workflow has no tags!