This workflow will allow the user to pick a data source and do some decision tree model learning on it. Valid data sources are an in-built SQLite database or the upload of either another SQLite DB file or a custom KNIME table file. The workflow will then analyse the table and offer the user too choose the target and independent variables. A (very simple) report will be generated and shown to the user.
Prints table with environment variables (output of 'env').
This workflow show the user different quickform which require its input. Afterwards they are shown in one table.
This workflow demos how a file can be upload to the KNIME Server and send via EMail. There are multiple sanity checks included. If no file was uploaded the user is asked to go back. If the sending of email fail another error message is shown. If everything works as expected a success message is output.
We here demonstrate how a date/time column can be filtered to contain only the last k month or week or days. On the webportal you chose the granularity and the number of those granularity you want to include in your analysis. Afterwards we filter them using the time difference (set to the selected granularity) and the rowfilter on the time difference value limited by the selected number of.
This workflow shows the usage of quickforms to make a configurable user experience. In the first node, the user gets asked to provide a number of rows and features. Using this information, a table containing random time series is generated. On the second page, the user can chose which of the generated time series columns should be shown in the line plot contained on the third page of the webportal.
This workflow demonstrates a usecase of the KNIME Webportal. Here a file can be uploaded and different parameters are chosen. The parameters provided here are the option that the first row of the table contains the column header and that the user can choose a column delimiter. Providing different parameters this workflow can be extended to read any kind of text based files.
This workflow creates a model selction on the webportal, where data analysts can choose a training and a test set between the datasets of 2006 to 2012, the target variable and the algorithms, which should be considered for the model comparison. The results are then displayed in a ROC Curve.
For further comparison two algorithms can be chosen to compare them via a Lift Chart, before the data analysts can decide for a final algorithm.