We performed a comparison between KNIME and Microsoft Azure Machine Learning Studio based on real PeerSpot user reviews.
Find out in this report how the two Data Science Platforms solutions compare in terms of features, pricing, service and support, easy of deployment, and ROI."It's very convenient to write your own algorithms in KNIME. You can write it in Java script or Python transcript."
"Usability, and organising workflows in very neat manner. Controlling workflow through variables is something amazing."
"I would rate the stability of KNIME a ten out of ten."
"The product is open-source and therefore free to use."
"The most valuable feature is the data wrangling, which is what I mainly use it for."
"From a user-friendliness perspective, it's a great tool."
"The solution is very easy to use"
"The ETL which helps me to collect, reformat, and load the data from multiple sources into one destination, a storage database."
"The drag-and-drop interface is good."
"I find Microsoft Azure Machine Learning Studio advantageous because it allows integration with Titan Scratch and offers an easy-to-use drag-and-drop menu for developing machine learning models."
"The most valuable feature is its compatibility with Tensorflow."
"It is a scalable solution…It is a pretty stable solution…The solution's initial setup process was pretty straightforward."
"Their web interface is good."
"The product is well organized. The thing is how we will get the models to work within our code. We have some suggestions there, but we want to gain more experience and be ready to answer that because we are currently working on this and don't have all the answers yet. The tool is well organized. What I am very happy about is the ease of deploying new resources. You can easily create your pipeline within minutes."
"Split dataset, variety of algorithms, visualizing the data, and drag and drop capability are the features I appreciate most."
"The solution is really scalable."
"When deploying models on a regular system, it works fine. However, when accuracy is a priority, hyperparameter tuning is necessary. Currently, KNIME doesn't have the best tools for this which they could improve in this area."
"The ability to handle large amounts of data and performance in processing need to be improved."
"The overall user experience feels unpolished. In particular: Data field type conversion is a real hassle, and date fields are a hassle; documentation is pretty poor; user community is average at best."
"KNIME is not scalable."
"The solution is inconvenient when it comes to wrangling data that includes multiple steps or features because each step or feature requires its own icon."
"They should look at other vendors like Alteryx that are more user friendly and modern."
"Though I can use KNIME in a 64-bit platform in the lab, it's missing some features. For example, from my laptop, I can use the image reader feature of KNIME. However, in the lab, the image reader node is missing."
"I'd like something that would make it easier to connect/parse websites, although I will fully admit that I'm not as proficient in KNIME as I would like to be, so it could be I'm just missing something."
"The initial setup time of the containers to run the experiment is a bit long."
"The AutoML feature is very basic and they should improve it by using a more robust algorithm."
"The data preparation capabilities need to be improved."
"In the Machine Learning Studio, particularly the Designer part, which is essentially Azure's demo designer, there is room for improvement. Many customers and users tend to switch to Microsoft Azure Multi-Joiners, which is a more basic version, but they do so internally. One area that could use enhancement is the process of connecting components. Currently, every time you want to connect a component, such as linking it to your storage or an instance like EC2, you have to input your username and password repeatedly. This can be quite cumbersome. Google, for instance, has made it more user-friendly by allowing easy access for connecting services within a workspace. In a workspace, you can set up various resources like storage, a database cluster, machine learning studio, and more. When connecting these services, there's no need to enter your username and password each time, making it a more efficient process. Another aspect to consider is the role of the designer, and they were to integrate a large language model to handle various tasks, it could significantly enhance the overall scalability and usability of the platform."
"There's room for improvement in terms of binding the integration with Azure DevOps."
"Microsoft Azure Machine Learning Studio could improve in providing more efficient and cost-effective access to its tools for companies like mine."
"Operability with R could be improved."
"When you use different Microsoft tools, there are different pricing metrics. It doesn't make sense. The pricing metrics are quire difficult to understand and should be either clarified or simplified. It would help us sell the solution to customers."
More Microsoft Azure Machine Learning Studio Pricing and Cost Advice →
KNIME is ranked 4th in Data Science Platforms with 51 reviews while Microsoft Azure Machine Learning Studio is ranked 2nd in Data Science Platforms with 54 reviews. KNIME is rated 8.2, while Microsoft Azure Machine Learning Studio is rated 7.6. The top reviewer of KNIME writes "A low-code platform that reduces data mining time by linking script". On the other hand, the top reviewer of Microsoft Azure Machine Learning Studio writes "Good support for Azure services in pipelines, but deploying outside of Azure is difficult". KNIME is most compared with RapidMiner, Microsoft Power BI, Alteryx, Dataiku and IBM SPSS Modeler, whereas Microsoft Azure Machine Learning Studio is most compared with Google Vertex AI, Databricks, Azure OpenAI, TensorFlow and IBM Watson Studio. See our KNIME vs. Microsoft Azure Machine Learning Studio report.
See our list of best Data Science Platforms vendors.
We monitor all Data Science Platforms reviews to prevent fraudulent reviews and keep review quality high. We do not post reviews by company employees or direct competitors. We validate each review for authenticity via cross-reference with LinkedIn, and personal follow-up with the reviewer when necessary.