Analytics
- Decision Tree to Rule Set (New node)
- Rule Handling (New node)
- Statistics measure as aggregation methods in GroupBy node
- Extended PMML Support (New node)
- Data Generation (New node)
- More Statistics Nodes (New set of nodes)
Tool Integration
- New MongoDB Integration (New set of nodes)
- Javascript Integration (New set of nodes)
- Extended JSON Processing (New set of nodes)
- XML XPath Interactive extraction (New node)
- Extended Python Integration (New node)
Utility Nodes
- CASE Switch (New node)
- Sleep / Pause / Timer (New node)
- GUI Improvements
KNIME Server & Automation
- REST interface
- WebPortal Branding
- Call Local / Remote Workflows (New nodes)
- KNIME Local Automation (New product)
Free Partner Extensions
- Streaming Execution DataFlow by Actian (New execution engine)
- See full list of changes in changelog file
Check out the video on YouTube about all new features in KNIME 2.12: https://www.youtube.com/watch?v=TD0iqWVVuA4 Analytics | |
Decision Tree to Rule SetThis new node converts (a single) Decision Tree model to a PMML-formatted Rule Set model and also to a data table containing the rules in a textual form. |
|
Rule HandlingThe Rule Engine (Dictionary) applies rules from a rule table to a data table. Rules follow the same syntax as in the Rule Engine node. The antecedents are stored in one column and the consequents in another column in the same rule table. The rule table is then input at the lower port of the node. Rule syntax is verified during configuration. A similar node, Rule Engine Variable (Dictionary), is available for working on flow variables. |
|
Statistics Measure as Aggregation Methods in GroupBy nodeStatistics Measures are now available as aggregation methods in the GroupBy, Pivoting, and Column Aggregator node: Percentile, Correlation, Kurtosis, Median Absolute Deviation, Second Moment, Skewness, and sum of logs/squares. |
|
Extended PMML SupportThe Missing Value node now exports a PMML model and can feed a Missing Value (Apply) node. It also supports more options such as "previous" or "linear interpolation". Another new node allows columns to be filtered that have more than a specified fraction of missing values. |
|
Data GenerationThere are two new nodes for numerical data generation under IO/Other/Modular Data Generation. Visual 2D Data Generator creates a two dimensional data set by manually adding/clicking points in a coordinate system. Quasi-Random Sequence Generator (Apache) generates a quasi-random sequence to cover all parts of a high dimensional space.
|
|
More Statistics NodesWe have started to add more statistics nodes (many more to come in future releases) under KNIME Labs/Statistics: Rank Correlation, Linear Discriminant Analysis, Cronbach Alpha and Standardized Cronbach Alpha, Odds and Risk Ratios, Wilcoxon Signed-Rank, Kruskal-Wallis test, Wilcoxon-Mann Whitney test. |
|
Tool Integration | |
New MongoDB IntegrationIn KNIME Labs, there is a new category that is dedicated to interacting with MongoDB. A number of nodes is available in order to read, remove, save, update, and write data from and to a MongoDB database. |
|
JavaScript IntegrationThere are many new Javascript D3-based interactive visualization nodes under KNIME Labs/Quickforms and Interactive Nodes/Views and Interactive Nodes. These nodes allow View Controls, among other settings, to be enabled during configuration. View Controls make the node view interactive allowing you to zoom, select, and make other changes on the fly. These nodes also produce a static image at the output port to be embedded in reports. |
|
Extended JSON ProcessingMany new additional nodes are now available in KNIME Labs for processing JSON structures: compose from table, JSON to table and vice versa, combine & write, and more. In addition, the JSON Reader node now supports remote URLs. | |
XML XPath Interactive ExtractionNew XPath node! This node shows the XML structure of the selected column. Syntax highlighting and interactivity help to build the correct XPath query. Double-clicking an element or an attribute in the XML structure automatically generates the corresponding XPath query. The Add XPath button adds it to the list of existing XPath queries. Automatically generated XPath queries can always be manually modified. | |
Extended Python IntegrationA few new nodes for Python Integration:
| |
Utility Nodes | |
CASE SwitchTwo new switch blocks: CASE Switch Variable and CASE Switch Model. These blocks behave similarly to the already existing CASE Switch (now renamed CASE Switch Data), but take / produce Flow Variables and a Model at the input / output port respectively. These nodes were originally developed by Vernalis and are currently maintained by KNIME. |
|
Sleep / Pause / TimerNew node called Wait ... This node waits either for a certain time, up to a certain time, or until a specific file event (such as file creation, modification or deletion) occurs. This node was originally developed by Vernalis and is currently maintained by KNIME |
|
GUI Improvements
|
|
Many more small improvements have been made under the hood - please refer to the changelog file.