Consultant's best (free) friend
Overall Satisfaction with KNIME Analytics Platform
I have used KNIME for advanced data analytics and experiments in the AI (machine learning) area. I have also used this platform for running client data analysis in sourcing and sales areas, including running of prediction models.
This is a framework that allows you to start with simple tasks and gradually increase the analysis complexity.
After going repeatedly through several data sources with tons of data, the painful part has always been preparing and transforming the raw data for analysis. This can be automated and the data acquisition model can be saved and run repeatedly, saving a lot of time. Data cleansing and blending of tables is easy here. It also supports formats as JSON, XML, a quite frequent format nowadays.
Above all the platform and community is wide with hundreds of add-on modules. Frequently, someone has already solved a similar task as you. Before trying to model anything from scratch, it is a good idea to skim through modules and hopefully you can find a good one to use. And finally, it supports simple as well as complex analytics, including AI algorithms.
This is a framework that allows you to start with simple tasks and gradually increase the analysis complexity.
After going repeatedly through several data sources with tons of data, the painful part has always been preparing and transforming the raw data for analysis. This can be automated and the data acquisition model can be saved and run repeatedly, saving a lot of time. Data cleansing and blending of tables is easy here. It also supports formats as JSON, XML, a quite frequent format nowadays.
Above all the platform and community is wide with hundreds of add-on modules. Frequently, someone has already solved a similar task as you. Before trying to model anything from scratch, it is a good idea to skim through modules and hopefully you can find a good one to use. And finally, it supports simple as well as complex analytics, including AI algorithms.
Pros
- Great UX interface, easy connection of data sources, good handling of the analytical model, easy to modify.
- It provides good level of control of what happens with your data in each step.
- Great tool from data preprocessing, from analysis to visualization.
- Great community and a lot of modules to reuse.
- Supports machine learning - it is easy to configure and run.
- It is Open Source!
- If you are familiar with Python, you can use this easy programming language to add additional functions to your analytical model.
Cons
- Automation - e.g. RapidMiner Studio provides a Turbo Prep function, where one can get to working on models more quickly (RapidMiner is not open source though)
- KNIME does not replace a regular reporting tool - it is not meant to. However, if I have already spent some time developing a data acquisition and analytical model, it would be nice to be able to deploy, for example, a monitoring or reporting module that would process data autonomously and react accordingly.
- Lowest TCO compared to other tools
- Accelerates analysis - the analysts can dedicate more time to analysis itself, not to data preparation
- KNIME is open source and thus brings the price advantage, compared to RapidMiner.
- RapidMiner uses some WOW features such as turbo prep function, good for quick small tasks.
- Both tools are excellent in data preparation and processing.
- RapidMiner is a bit easier to install and setup.
- Both tools are not strong on security around the framework, which can become important when dealing with confidential data.
- Both tools are for ad/hoc analysis and report creation. It would be useful to be able to generate a self-standing monitoring module, that evaluates data according to sophisticated rules and "does something" when the tool finds a value out of boundaries or above a certain threshold.
Comments
Please log in to join the conversation