The ML analysis process involves executing ML analyses, managing and editing them as needed, and sharing the results with others. By configuring models, queries, and filters, you can analyze data and generate predictions.
- Use the search option to look for any existing ML analysis.
- Use the Filter by Datasource option to filter the ML analysis based on the datasource.
- To switch the view between Card View and List View, click on the respective icons near to the search bar.
- To create a New Folder in the ML analysis, click on the Folder icon.
- To download the list of all existing ML analysis, click on the Download icon. It delivers the details on (• Entity Id Name • Description • Content Tags • Created By • Created By User • Created On • Folder • Datasource Id • Datasource Name • Is Folder Public)
Execute Machine Learning Model
Perform the ML model analysis
To perform ML analysis in Infoveave, follow these steps:
- Access the ML Analysis section by clicking on Analysis ML Analysis.
- You will see a list of all ML analysis, including those created by you or shared with you.
- To create a new analysis, click on “New ML Analysis”. The ML Analysis dialog box will appear with three tabs:
- Analysis Builder
- Columns Mapper
- Form Data
- Analysis Builder:
- In the Analysis Builder tab, assign a name to the analysis and select an ML model from the drop-down list. These models were created previously in the ML Modeler.
- Choose the prediction source from, Query, File or Form Data.
- If selecting the query option, choose a specific query from the drop-down list.
- If selecting form data, configure the column details under the Form Data tab.
- If selecting file, choose the file from your local folder.
- Apply filters based on dimensions if applicable.
- Columns Mapper:
- In the Columns Mapper tab, you have the option to map each column in the dataset.
- If any columns have different names in the train and test datasets, you can map them by selecting the appropriate columns from the drop-down lists.
- Check the “Same as Source” checkbox to map all columns as the same as the data source columns.
- Click Save to save the ML Analysis.
- You have an option to map each column available in a dataset under Columns Mapper tab.
- Assume that in a dataset, if a column has a different name in the train & test dataset. Then, you can map source column to destination column by selecting a column from the drop-down list.
- Check Same as Source checkbox, if you want to map all columns as same as datasource columns. In this case, the drop-down list of destination columns will get disabled.
- Form Data:
- Choose the “Form Data” option for making predictions.
- The Form Data tab will display the columns selected over the query.
- Review the columns listed, from the query.
- Configure the column details for each field based on your analysis requirements.
- Make sure to provide the necessary information or mappings for each column, if applicable.
- Apply any additional filters or dimensions, if applicable, to further refine the analysis.
- Click “Save” to proceed to the next step.
- Click on Execute to run the ML analysis.