RapidMiner Reviews

the user interface and experience for sure. it's so easy to use especially for a noob like me. it's very intuitive and you can grasp the basics within like 5 minutes of you using the software. also there are lots of ways which i can use the graphically represent the data that i work on, also the graphs that i can make look really clean and professional. and i'm not a tech person but it feels like it puts less pressure on my dingy old laptop that excel. like if i use excel to even load a sheet with more than 5000 columns it lags and freezes. RapidMiner did it in under 30 seconds and when i was like using it i never heard my laptop fan coming on or any weird noises. so that made me trust it a lot.

nothing as of yet. the only problem i faced was due to me being new to the software, not anything about the software itself.

just buy it man. it'll save your life. it lets you know clean a sheet as well if some noob has messed it up for you.

okay so basically i'm a student right now, but i used to work as a fintech marketer. a tool like RapidMiner would have been priceless to me when i was actually going about running digital campaigns to get people to invest. and not just in terms of drawing trends and figuring out cluster patterns amongst large databases, but also about campaign data itself. right now as a student, i'm using it to show my professors the correlations between different variables in emerging markets, and to visually represent those particular relationships.

I have over 20 years as a computer science BS programming in all and every language that was corporate accepted. I am a leader/influencer/specialist in data and analytics and have been using RapidMiner since the command line version. RapidMiner can keep up with my analytic thoughts as compared to other tools which require me to pause and code to get my idea expressed. Love this tool.

My one critique is that it is challenging to do very large data on my desktop (20 years monthly linear regressions for about 2500 global companies with about 47 variables) SAS can do this in seconds. A second critique is that in KINIME there are many post model interpreting objects to easily perform analysis. This is improving in the current versions and I look forward to more improvements here.

Even if the company does not use the sever, etc for implementation in production...this tool for a data analytic is like Excel for those working with numbers and data (like an accountant). It sits on the desktop and can be used to model many thoughts and ideas that can then be transposed into the corporate standard for predictive modeling. I have tried most all/many tools on the gartner magic quadrant and RapidMiner is by far the best and most usable tool of all. They each have strengths but as an overall experience for rapid analytic modeling RapidMiner is the tool.

I am a researcher so I am doing investment modeling (I am a quant) that is not being done by others. I use this tool to flex my skills in a way that is not normal. My benefits are that it can let me build out prototypes of research that I then have to convert to SQL & SAS. I am using memSQL and really want RapidMiner to connect and do the heavy lifting on that machine (it has NVIDA memory) because I am doing millions of row modeling. I have an enterprise desktop version on my desk at work. I also teach Data Analytics and write graduate coursework for SNHU and drive my students to use RapidMiner vs R/Rattle. It is going well. Especially the Auto Model in the coursework.

Auto Model allows me to quickly understand the best modelling to use on my data without having to build the workflows manually. This has allowed me to apply more complex models to my data then I would have otherwise felt comfortable/confident doing. The ability for someone like me (who is primarily a data visualization artist) to quickly and easily model complex data and do correlations that are understandable is AMAZING! Also the company themselves are super fun and helpful anytime I've had an issue or question. You can't fake the funk on being helpful as a company!

The complexity of the program is daunting to newcomers which is why Turbo Prep and Auto Model are LIFESAVERS! They offer many tutorials to overcome this daunting nature but the sheer size of the program means you could spend all day watching tutorials.

Don't be frightened off by the complexity of the program! Ask questions on Twitter or in their community and you'll get an answer QUICK!

Currently I'm using RapidMiner for some social good projects I'm working on to re-examine previously analyzed data, applying newer and more up-to-date statistical models which allow me to discover new insights in older data. Allowing me to find more ways to help the communities I'm involved with stay safe if it's looking at traffic collisions, roller derby injury results, or gender and relationship studies to disprove stereotypes. Using data to help with public good projects is easily the most rewarding part of my career as a data visualization artist and RapidMiner is a crucial part of that analytical workflow.

RapidMiner is easy to use. With a inside application tutorial, you can quickly start using for data mining.

It's easy to work with machine learning algorithms with user friendly interface.

In addition, RapidMiner has most of the algorithm for data mining. Beside, it can extended programming with Python or R.

One of the things I liked most is the documentation that can be found inside the application which guides me through implementing the data mining solution.

I don't have anything special to dislike. Except RapidMiner is too slow in comparison with other software and I hope it can improve speed more.

This is a great tool for data scientists who want a user friendly tool for mining their data.

I am using Rapidminer in my research about drug-drug interaction prediction. I am also using it in teaching my students data mining class.

Corporate leaders, at various managerial levels, have been challenged by fears that any day soon, a competitor of inferior technical capacity would jump ahead from behind. They were amazed at the fast and easy practice when we demonstrated the RapidMiner platform. Some of them have also suffered from ill-fated projects in the past, perfectly aware of the negative impact due to slow, error-prone, and delayed data preparation, model training, and deployment. Understanding its advantages, they too realized that others can easily outrun them quickly with RapidMiner.

Granted, companies are looking for vertical solutions aiming directly at their specific applications while RapidMiner, albeit powerful, is a platform. In addition to current communication by RapidMiner and the enthusiastic community, efforts must be spent by both sides, the customers and the solution providers like us, to practice, solve problems, and build up confidence so that gaps can be bridged.

It is rather idealistic to wish for a vertical solutions aiming directly at specific applications. You may have concerns that RapidMiner looks powerful but is still a few steps short. You are not alone but others have used RapidMiner to complete data science projects in weeks if not in days. With patience and determination to become a good problem solver, RapidMIner will bring you far ahead of competitors in a short period of time.

Data science is after all an art of problem solving. We have encountered problems such as, in predicting disgruntled customer, “can we generate some statistics for the past calls that belong to the same customer of the incoming call?” and, in predicting the lab-generated quality measure in a chemical plant, “can we shift each and every dependent variable according to their respective time lags?” Yes, RapidMiner can support them and thus incorporate the steps in a RapidMiner process. Such problems are however slightly beyond the basic practice of RapidMiner.

Rapidminer is one of those tools that delivers a wide range of Data Science capabilities. If you are savvy with your data and conceptual understanding of the field you will immediately appreciate the abstraction layer that provides a wide range of tasks across the CRISP-DM lifecycle. It is a great tool that accelerates tackling the Data Science questions at hand fast. One of the greatest features is the Auto Model feature. It helps with assessing a whole bunch of algorithms (including Deep Learning) versus a given data set at once. Which enables users to best decide beforehand which algorithm would be the best answer to for their questions.

The addition within the new version of the turbo data preparation also transforms a lot within the CRISP-DM development lifecycle. It literally saves 80% of the time wasted on preparing the data through many running both PCA and normalization. The tool also addresses the problems of multicollinearity and missing value imputation usually diminishes the performance of the models.

Quite a complete Data Science tool !

Visual graphs from the different algorithms within the tool

Artificial Intelligence research work

Entorno de trabajo cómodo, e intuitivo, con ayuda durante el desarrollo del proceso

Actualizaciones apropiadas, con mejoras, como la posibilidad de automodelización

Cantidad de algoritmos para modelar, agrupados por características...

Buenos gráficos, y opciones para visualizar el resultado del proceso...

Facilidad de parametrizar los diferentes algoritmos...

Visualización del proceso de minería de datos....

El contenido de la ayuda es apropiada....

Pocas opciones en los gráficos de tres dimensiones, como por ejemplo, el tamaño de los puntos, que son pequeños, si fueran mas grandes se podrían visualizar mejor...

Me gustaría que los gráficos en tres dimensiones se puedan mover sobre sus ejes automáticamente, durante un lapso de por ejemplo, cinco segundos, y poder grabar la secuencia para exportarla a un documento, como si fuera un video, que permita ver por ejemplo, la composición de los clusters...

Me gustaría la versión en castellano, para comprender mejor el contenido de la ayuda...

Estaría bueno que la versión educacional se pueda extender a los alumnos y los profesores, y poder realizar minería de datos en equipo de trabajo, tal como en un curso, o en una materia de posgrado o doctorado, con mejores opciones de almacenamiento en la nube.

Me gustaría que este en castellano

Estoy usando la aplicación para estudiar los algoritmos, base para mi tesis doctoral de Ciencias Económicas, mención Contable. En principio me interesan los modelos para detectar fraude, y poder minar datos en Informes Financieros, o dataset impositivos.

- Intuitive interface, robust software operation

- Almost every imaginable data input format is supported

- Constant development and adding new features, big and small. Super fast bug fixes.

- Huge coverage of all possible machine learning algorithms, I think ~98% of all imaginable data science use cases can be covered with RapidMiner only

- Scalability and deployment capabilities: from studio to server to deployed models in very short time.

- Super quality of educational materials provided, adding new materials all the time

- Very thorough documentation on product capabilities and features

- Responsive and active community always willing to help

- Poor graphic plotting possibilities, very little option to configure plots

- Not all runtime error messages are intuitive enough

- Process and repository handling could have used more transparent versioning control

Don't expect 'magic': you still have to understand basics of machine learning, as well as spend some time learning how to use the software efficiently. If possible, RM training course is highly recommended. Once this is done, you'll experience full benefits of this great product.

- Fraud detection modelling

- Customers segmentation

- Text analytics

- Data mining

Main benefit is the speed of at least building prototypes - you can get a model ready in some 30 minutes, including all the data preparations.

I deal with messy data from many sources. I need a tool that helps me pull together and utilize data in this kind of environment as efficiently as possible. The company seems to be very responsive to customer needs and keeps adding functionality that is practical for the user. I'm really impressed with the interface as well.

It's a little difficult to share RapidMiner files/processes with the other staff member in my office. Our budget will never allow for a full on server type sharing product. So I guess I feel like RapidMiner's products are heavily geared towards corporate uses, and sort of fail to meet the needs of small, education, and non profit organizations. Everyone has data management needs in the modern world; but lots of organizations aren't about using data for generating profit. We need tools too.

I especially love RapidMiner for the data cleaning and preparation processes that I have to repeat either annually or each semester (I work at a college). I also often use RapidMiner to validate my data or look for potential errors.

- The graphical modelling workflow: it's not only easy to design and build a machine learning model, but, even more importantly, to explain it to internal customers (your boss, your colleague, end user, factory staff, you name it).

- The Automodel feature: it is enormously handy in setting up a quick data analysis. It guides the user from the begin till the end. It provides very useful metrics about the imported data, and builds a couple of pre-defined models for you.

- The plotting and visualization is really crude. The customization of the graphs is hardly possible. Exporting the graphs (to PowerPoint for example) is (afaik) not trivial - I had to use screen clipping for the lack of a better alternative.

- Apparently there is no tool for exploratory data analysis.

- The execution of complex workflows is serial, even though there are obvious parallelization opportunities.

I used RapidMiner for prototyping machine learning workflows. For me, it shortened the time needed for setting up a decent ML predictive model from ~2 days to half a day. This includes evaluating several model prototypes and getting the insights what model would work best with my data.

NO BLACK BOX concept. I have used other data science tools, but my concern was how these tools determine optimal values for algorithm and what are these value. But in RapidMiner I can self tweak and automatically optimize the parameter values as well. This is just one of the many benefits I had from No Black Box concept.

Intuitive GUI is another USP of RapidMiner Studio for beginners and experts. It has a help section with detailed documentation and quick tutorials which is very informative to know the logic behind every operator. I have learnt a lot of concepts from there.

RapidMiner Server needs to be redesign and more dynamic functionalities can be added in Web App.

RapidMiner Studio should have multi-process tab and github integration(just a suggestion).

I have been solving diverse problems using RapidMiner. I've tried algorithms on clustering, anomaly and outlier detection, classification, regression, keras, Text Mining (I've explored Rosette, Aylien text, meaningclous, edda etc). I have also been using RapidMiner for ETL jobs.

In terms of benefits, it has reduced my development time by many folds. Coding the same jobs in python used to take so long and then deploying it into production was again a challenge.

But RapidMiner is super easy to build solutions from end to end.

I have a two-core i3 that I run RapidMiner on. I don't work with millions of data but I do teach. The software is stable to any query you may have and only consumes the necessary resources. That means that your code is designed only to cover my query and has no "hidden" routines that are consuming my CPU or memory.

The community behind Rapidminer is a guarantee that no matter what problem you get into, you will find help and a possible solution.

Plugins help me to experiment with different approaches to address a problem, so for research, RapidMiner is a very useful toolbox.

Without much effort I have obtained an education account and this has allowed me to use more data than I used initially.

While the interface is intuitive (it reminds me of the psychomotor process used in the MIT Scratch application), the learning curve is pronounced. I've been using this tool for a couple of years now, and I still don't feel safe teaching my students about RapidMiner.

The problem is that there is no documentation in Spanish and only a few videos in my language. The problem is not the amount of material, but the depth of the content. Learning the basics of Rapidminer is easy, finding the details of how to use the plugin parameters is difficult, even in the English material.

I would like the community to meet in a VBulletin-type forum, the current tool is not professional enough.

RapidMiner scripting through Python is simpler than the R programming language, although the programming is still complex (I am a programmer).

Useful for data recovery, processing and analysis. Ideal for entering the world of Data Science. Useful for those who do Text or Data Mining.

The professionals who come to my class come with needs to acquire skills in structuring, processing and analyzing financial data. Excel or PowerBI are not enough for some of the problems being addressed.

Some of them arrive with some queries that they can't make in IBM SAP or find very complicated in that tool.

In the Business Intelligence Diploma, when we have to talk about Data Mining, I give them guidelines to enter the world of RapidMiner.

The graphical user interface is very intuitive and RapidMiner is very easy to start with.

It offers good and meaningful default values for beginners in every operator.

Since you are able to see what you are doing, it is quite easy for beginners to understand

the science behind.

The computation speed is really fine even on notebooks (when trying out to sketch a new process).

RapidStudio charts are looking a bit old fashioned and simple (compared to BI solutions such as PowerBI or Tableau).

Configuration of axes and labelling formats is not possible in the simple charting modes and the complex charts are not easy to configure.

You will be able to get data easily into RapidMiner as well as first useful results.

The RapidMiner Book is great and offers a good starting point for analysing with RapidMiner.

In combination with a other books (such as "Data Mining for the masses" from Matthew North as well as

"Predictive Analytics and Data Mining" from Deshpande and Kotu), soon you will know enough to help yourself and tackle every data problem.

I adore Ingos youtube channel and he has got talent explaining things easy.

That's a real valuable ressource for Data Scientists and students alike.

Since I am teaching, I am trying to cover everything from clustering, classification,

association rule mining and regression. Some of my students have used RapidMiner Studio for Text Mining purposes as well and where obtaining good results (within their thesis e.g.).

The UI is very intuitive, it's easy to play around with advanced machine learning algorithms with the simplicity of a drag and drop, flow-based interface.

It's easy to integrate RapidMiner with and ETL process, as it can connects to several DBMS and also Cloud Storage services, and it is also easy to integrate it with BI Tools, as it can output data to DBs or even expose data through webservices (OData) by using Rapid Miner Server.

One of thing I liked the most is the impressive quantity and quality of the documentation, that can be found both inside the application and in the website. It's not just a getting started documentation, it really delves into the details of each algorithms and gives you a non negligible free know-how; there is also a considerable number of tutorials and examples directly accessible from the documentation items. On top of that, there is a very effective series of video training available on the website, which guides you through implementing the major steps of a Data Analysis solution.

The RapidStudio Server UI it's a bit old fashioned.

It has a very good customer support and they offer free webinars on how to use the application.

Ingo, the company founder, also has a YouTube channel where he effectively explains the basic idea behind the most important ML algorithms and why to use them.

I'm trying several techniques for outliers detection, machine learning algorithms offers several way to do that and RapidMiner is an easy and ready to use solution that implements those algorithms.

Comparative analysis of results from all possible models from their figures of merit in terms of errors from the prediction in the shortest possible time

Does not generate a model diagram or graphics for Deep Learning models also Predictive Model equations

Consider Rapidminer to quickly determine the interacting factor or attributes that can contribute to a predictive model and then dig deeper into the best model it suggests.

The determination of interacting attributes from agronomic factors or attributes in Pineapple fruit growing to be able to predict Yield and Vegetation Health. The complexity of agricultural processes that are measurable in terms of attributes or factors can now be modeled way beyond the usual linear regression and statistical design of experiments.

The simple useage and the possibility to learn all necessary aspects in a short amount of time so you can do professional data science very fast and easy. The frontend shows all essential parts on one view and you can customize the view exactly to your needs to support your personal preferences while working. The flexible design of a process throgh the Operators that are exactly described right within the Software and even linked to a use case showing the function of each Operator.

The large amount of RAM used to handle large datasets (e.g. to join them) and no possibility (i know) to have an alternative to work around this Problem and the vizualisation engine feels a bit slow slow sometimes when handling large datasets.

I use Rapidminer to do Predictive Analytics processes as well as Research with Text Mining proceses in the area of Marketing and Sales. I get a lot of additional insights to use and have a scoring for potential customers to evaluate the best possible customers on the market.

I know a lot of Data Science/Data mining/Stats software, but RapidMiner is the best at this moment. Easy to use, Cost is low, has most of the algorithm/stats, Collaboration with Python/R. You can try any models by a free license (running up to 10,000 records). When you try Turbo Prep and Auto Model, you will be enchanted by RapidMiner.

RapidMiner should open up new functions such as Turbo Prep and Auto Model to Free users

I'm Positive very much.

It gives data science evidence to my business reports. Predict demand from time series data. Find the correlation from many variables. Find the continuity of demand from many variables. Find the trends of demand from many variables.

The options for choosing between different machine learning algorithms is the key feature for the RapidMiner software, the extensive documentation about each operator is very helpful for the beginners. The new option of data prep is similar to the data prep options available for various business intelligence tools which had streamlined my pipeline and made it more efficient.

The only downside about the RapidMiner is the data visualization part, the basic charts/visualization is easy to customize but when you dig deep and perform advanced visualization that's where the entire process becomes very complicated, I have tried numerous times the advanced features after reading the documentation but couldn't figure out the parameters and the customizing the graphs options.

This product is apt for predictive analytics and it offers a wide range of machine learning algorithms from unsupervised, supervised, and association analysis and the option of parameter tuning helps to cut out the time spent on programming languages while this product simplifies the data ingestion process and have the option to connect to various data sources. The documentation about each operator is offered with an explanation which is resourceful.

1. The option of ingesting data from various data sources

2. The process of preprocessing the data or data munging is streamlined.

3. The option of choosing between various machine learning algorithms is helpful to compare the accuracy rate and efficiency between the forecasting models.

4. The option of time series modeling and forecasting.

With Rapidminer, I don't have to look up how to code something. My coding isn't bad at all, but even with that, with so many available frameworks and different API documentation, it becomes really hard to figure out just the right lines of codes to accomplish the task.

Rapidminer helps me design the models in ways that are intuitive and I literally just need to make some connections and viola! it's done.

Every tool has some problems, so does Rapidminer. While being abstracted from all the happenings inside the "operators" has its pros, if you don't have any idea what it actually does, making the right tuning really becomes difficult. Documentation that keeps in mind that maybe the user doesn't know anything about the operator or what is does and explanation to the simplest level is needed. I know this might be a tough thing to do, but if I had to "dislike" something in Rapidminer, this would be it.

I am a student and haven't really solved any business problems per se. What I have solved are problems on various datasets that are available through various competitions or are open-source.

The benefits Rapidminer provides in EDA is HUGE!! I can get a general idea from the 1st step itself and that's awesome. Also, tinkering with the models is quick and easy.

RapidMiner is a comprehensive and intuitive structure for managing end-to-end machine learning problems. Data can be easily aggregated from multiple sources and prepared thru standard reusable operators and processes. Applying multiple methods to gain insights into useful patterns is easily performed\learned in a short time without having to master Python, R, and other frameworks.

It'd be helpful to provide more transparency in the underlying "math" for those wanting to learn how calculations are performed. Some of this is easy to get to, but some of the methods provide results without the ability to drill into the underlying calculations. Maybe there could be links to open resources like Coursera, Khan Academy, MIT, etc. where excellent examples beyond the RapidMiner documentation and tutorials are provided.

We're using RapidMiner to measure the susceptibility of infrastructure failure (water systems, energy, roads, etc.)

Pretty much every possible information input position is upheld. Instinctive interface, powerful programming activity. Consistent advancement and including new highlights, of all shapes and sizes. Very quick bug fixes. Gigantic inclusion of all conceivable AI calculations, I think ~98% about every single possible datum science use cases can be secured with RapidMiner as it were. Versatility and arrangement abilities: from studio to server to sent models in exceptionally brief time. Excessively nature of instructive materials gave, including new materials constantly. Extremely exhaustive documentation on item abilities and highlights. Responsive and dynamic network continually eager to help.

Poor realistic plotting potential outcomes, next to no alternative to arrange plots. It's somewhat hard to share RapidMiner records/forms with the other staff part in my office.

On the off chance that conceivable, RM instructional class is profoundly suggested. When this is done, you'll experience full advantages of this extraordinary item. It is valuable for information recuperation, preparing and examination. Perfect for entering the universe of Data Science. Valuable for the individuals who do Text or Data Mining. It has a decent client backing and they offer free online classes on the best way to utilize the application.

RapidMiner is a simple and prepared to utilize arrangement that actualizes generally calculations. Fundamental advantage is the speed of in any event fabricating models - you can prepare a model in nearly 30 minutes, including every one of the information arrangements. Extortion identification demonstrating. Clients division. Content examination. Information mining

I just love that you can do what seems to be everything. It is easy to get started with the AutoModel feature and build up your experience with both data science and the product. I like the concept of putting feature blocks together and also the flexibility with the marketspace and integrations with python and different machine learning libraries.

I'm struggling to find something that I actually disliked. For me it was a great experience and the struggles I had was mainly due to my own impatience that I wanted to learn everything at the same time. I think its important to start small and finding your own pace during the learning process, because it can be a little overwhelming at first.

I used the RapidMiner for analyzing network related traffic in order to extract and analyze threat intelligence. For me the AutoModel was a game changer in order to easily create many different models and attribute selections.

Graphical User Interface and wide range of available operators make common data mining operations easy to understand, accessible and allows building of complex operation above them.

Some of the operators such as Outlier Detection take lots and lots of time in computation, for a month consumption data of electricity energy meter the operator took 10 minutes before I aborted the operation and I have yet to find outliers in my data using RapidMiner.

RapidMiner is a powerful tool for implementation of data mining techniques and provides a wide range of operators, also GUI is easy to use and has a great online support community.

Pakistan is an developing country and there is a gap in electricity generation and demand. I am using electricity data to optimally minimize this gap by shedding load uniformly across customers without disturbing their normal consumption pattern and comfort level as "Thesis Work".

Modular construction of a model. Taking it one action at the time, that way you do not have to redo the whole module. It is also a lot easier to find where you need to make a change.

I cannot think of anything I dislike about Rapidminer at the moment.

If you are serious about Data Analysis, then you need to try Rapidminer. I introduce Rapidminer to all my Lean Six Sigma Green Belts and Black Belts, and everyone of them has indicated that they were seriously impressed.

In the past I used SPSS modeler, but since IBM has bought this software I have lost track. Rapidminer uses similar modular methods and I am still in the process of becoming more familiar with it all, but it definitely has a lot better educational support.

RapidMiner is very easy to use. With a little bit tutorial, a beginner can quickly start learning data mining by using RapidMiner. This tool is written in Java, which means that it can work on any platforms. Its basic version is free, and for education purpose the fully function version is available for free, too. I adopt this tool as a major data mining tool in my Introduction to Data Mining course. It is best for students with little data mining knowledge.

In addition, the documentation is great. Users can read documents on the right-bottom when an operator is selected. This makes it so handy to learn the details about an operator

So far I do not have anything that I dislike.

this is a great software that I would recommend.

1. using as a major tool for teaching students.

2. I also begin to use it for my own data mining research

RapidMiner allows you to utilize the discoveries of the statistical/mathematical/programming communities with little prior knowledge of theory and no experience with coding. The RapidMiner community provides educational videos as well as face-to-face workshops and webinars. On top of that, there are tutorials in product that help new users adjust quickly to the program and to the field.

The company promotes education at its own expense and is doing great work for the business/data science communities. Some webinars/videos that they provide are business orientated while some are theory leaning. The only thing that I dislike is the amount of videos (there are hundreds) to watch and so little time to do so.

Try it out for yourself, it's free! The tutorials take less than a few minutes and you'll catch on quickly, I guarantee it!

Programming is allows flexibility at the cost of time. RapidMiner saves that time spent configuring code and allows you to link other languages such as R to restore that flexibility.

It makes running the algorithms easy even with big data.

Organizing the dataset files in such a way you see everything better.

Very helpul and very easy to use, just basic knowledge in ML and RapidMiner does the rest.

The weka extension in RapidMiner iq very helpful I couldn't work with the application of Weka but due to RapidMiner I did.

Actually I didn't dislike anything like really, I love that tool.

Just maybe I have some recommendations, it would be very helpful if each time you select an algorithm, maybe a video could pop up describing how to use the algorithm which other algorithms you need with it. It will be very helpful.

I have made couple of researches, and they are all related with ML and big data, so RapidMiner is the number one tool to use in this case.

Complex analyses of complex data is made simple in a drag&drop style, presented in layers of abstraction and well documented in understandable terms. This makes an analysis reproducible and easily transferable to another application. In contrast to scripts that cannot be used without its developer, RapidMiner processes are at hand to everybody, instantly.

It may happen that you are not so happy with the version that is not for cloud computing. Ok, buy the one with Radoop and all support!

Try the introductory case processes and play! When going productive, book training and consulting by RapidMiner to gain the full benefit.

Resource saving in production processes is a large field ranging from anomaly detection based on time series over quality prediction based on sensor/machine data from the production process and finally to prescriptive analysis which controls the production through multi objective optimisation.

We use the Rapid Miner software as a principal software tool in the educational process of future MSc. degree graduates in Control Engineering, courses in Artificial intelligence. The aspects we appreciate most are:

- Availability of educational version on 1-year basis

- A wide range of provided methods

- Ease and real rapidity of the use (worthwhile of the tool name)

- Permanent software development

- Good acceptance from the side of higher-education students

- Good and rapid communication with provider representatives

I do not have anything special to dislike. The only thing (just to say something) - when moving from older to newer versions, change of the names of operator grouping have caused some confusions to me. However, maybe an idea for future how to help future customers: in many times unexperinced user may be desoriented by amount of available methods - what about providing an in-built questionaire-like knowledge-based system that would recommend a particular method (or a group of methods) proper for the given problem ... based on questions like: format of input data (numeric, nominal, both), label type (numeric, polynomial, binominal), missing data (yes,no), type of problem (classification, regression, rules), output model (cluster, centroid, hierarchical, prototypes,...)?

Just try and you will see.

At the moment the tool is used to prepare (educate) future IT professionals (in future mostly employed in electrical engineering industry) - data transformations, classification and regression problems, etc.

Definitely the ease of use. The interface makes it very approachable to people without a technical background. Have drop downs/set options for operators is a lot better than digging through code documentation.

As with all software, there are still a few bugs (but the support team moves to address them quickly). I do miss using find-replace within my R and SAS code for quickly changing processes to a new problem. I find that I have to go into several operators to fix things in them. Ultimately, de-bugging can take some time. Also, I am still highly reliant on RapidMiner support since the user community is a little smaller than something like R.

Build out an in-house training program with tiered internal support.

Right now we are using the text mining and similarity operators to identify duplicate or near duplicate database records. Also, we are doing time series prediction for daily call volumes in our call center (but another group is working on that) Also using it for process improvement, shifting manual queries from databases and excel into a single workflow.

GUI is easy to use. Unlimited data size. Speed is very fast. and the most thing I like is the support. License came quickly after I fill in the form. And I can use a free education version in 1 year. That help me so much

I don't any thing to dislike. But I hope it can improve speed more.

That is very good for data mining. Although I am a newbie and I don't use all of tool of rapid miner, it is still perfect, atleast i think.

I use RapidMiner to train Self-organizing map for my research. That is a machine learning method in drug discovery. It help me reduce the size of my data, and i can find compounds which is similar to my active compounds.

About RapidMiner I like best the graphical modelling workflow. For educational purpose it is easy to play around with different algorithms and learn their function. The students are able to see what they are doing and can understand algorithms for data analysis.

I don't like that the description of complex operators usually contains in only basic informations.

I recommend RapidMiner to people who need a program for data analysis and processing. It is also a great aid for teaching both high- and low-level courses.

I am using RapidMiner for educational purposes. The main topics are text mining, text classification and sentiment analysis.

Rapidminer allows you to build a rationale fluidly. The way the tool was designed makes the whole focus on building powerful analytical models without worrying about specific points. As you build an analytical model and are faced with the need to perform some transformation or correction, you'll find the right operator on RapidMiner to solve the problem. It's amazing how Rapidminer keeps its efforts in the right spot, without going through the technical implementation line that is time consuming and without allowing you to develop a simple and wrong model. The flow of the tool causes an immersion in understanding aspects of each operator used and give you security of the model that was created. All this with a good performance for large databases. For me, working with market intelligence is perfect, keeping my focus on finding the best results. At the same time that you use RapidMiner, you develop a learning cycle that transcends the tool and goes into questions about how the various algorithms work. You are asking the right questions and increasing your power of understanding.

Although it does not seem to be one of its focuses, it would be interesting if RapidMiner had more options for visualizing data and that in some way facilitated the construction of reports, without the need of other tools, such as Excel or Tableau.

A great tool.

I have mainly used for the construction of analytical models focused on analysis of unstructured data, mainly of social media. I have currently used RapidMiner to analyze sales data and other quantitative data, such as demand forecast data.

I have used this tool since I started data analytics in college and found it relatively intuitive to use even with little mining/ analytics experience. I like the speed at which I can build processes- and the ease at which I can test and change out elements. I also like the five minutes with Ingo series!

Occasionally the documentation can get a little difficult to understand.

Versatile tool that goes beyond basic etl and is constantly improving

Churn models. Text Mining. Sentiment Analysis. Data cleaning.

Always amazed at the speed at which data can be tamed and turned into insight. Sometimes there are hiccups and it takes longer. Most of the time, however, reporting the findings takes the bulk of the day.

I like intuitive user interface. Because of it, novice users quickly get a grip on basic machine learning steps. A lot of time spent when performing a machine learning task is usually spent with data pre-processing and trying to understand, what data means. Rapidminer offers a plethora of possibilities to import the data from various sources and formats, as well as some really powerful data visualization tools. All major ML models are also included. It is also very nice to have free academic licence available - for students and instructors. Academic licence is easily and quickly obtainable and the process does not require lengthy bureaucratic procedure.

I personally dislike that some metrics for performance of the models are missing. I have been unable to find adjusted R squared parameter for evaluation of multiple linear regression models. I could also benefit from additional metrics for feature relevance. For example i could benefit from standardized multiple linear regression coefficients.

Look at their tutorials. They provide quick start into machine learning.

If you are entitled for academic licence - it is free. When you will need it to support a business case, in my opinion the licence price is reasonable.

I am using Rapidminer in lab classes. I am also using it to rapidly test and evaluate research related ideas.

Rapid Miner is self explanatory and very easy to use tool. Following the videos on the website is enough. Once we started using the tool, the vendor has reached out to us to know if we are facing any issues and helped in understanding the tool. We were able to use the tool for data cleaning, data analysis, data modeling.

Rapid miner do not have wide variety of options to visualize the data as charts and graphs. I would like to see more graph options included.

We used the tool on smaller data sets. Try the tool with larger data sets to see the performance.

We are trying to use the tool to study the churning, performance and clustering of clients. The studies resulted in lot of insights which we can use.

RapidMiner is a predictive data analytics software but also offers capabilities such as AI/ML program. Functionally, it provides thet same capability as using programming languages, but utilizing a graphical user interface to develop process models. This utilizes "parameterized" process blocks, the same as if they were programmed.

RapidMiner has a community edition for 10,000 rows of data to process for free on the desktop. Advanced paid editions are capable of automated process generation, and installation into cloud environments such as Amazon Web Services. The Server edition enables the process automation. There is a marketplace that allows you to add extensions to the baseline package.

RapidMiner is a great way to get into concepts such as natural language processing, time series data analysis, web mining, k-means analysis, data clustering, recommendation generation and other capabilities. It also has native support for R and Python, as well as their libraries.

Any tool in the AI/ML space will require learning of the concepts and how they are are applied in a computational tool. This means learning about vector and matrix math, and general statistical capabilities to be productive sooner.

I would recommend learning about the AI/ML concepts from places such as Coursera or Udemy. Then to RapidMiner's site and YouTube channel to learn in greater detail, as well as the mechanics and operations of RapidMiner itself.

We will be using RapidMiner for natural language processing in healthcare, 2D graphical brain maps, analysis of billing fraud, and reduction of errors in medical coding.

when it comes to science, you need to keep it on track with a rigorous framework , even though you are offering an easy to use platform. They actually deliver. The UI is intuitive, which is a plus for the user, but still there is not compromise on the fact that users have to know the how, what and why of their research, analysis or mining.

There are plenty of information, tutorials and resources (videos on youtube, for example).

quick and easy: pricing system, which unfortunately is a common feature with so many platform out there.

so far, I have experience with a peculiar extension, related to process mining. Clients directly won't get it, but this extension let you perform several analysis reducing mistakes and delivering reliable insights (if you design them in the right way).

Even though it is not my primary tool, I would recommend to spend time on it.

RM provides an easy access to the world of predictive analytics. In my view, it is as useful for that purpose as - say - MS Access for SQL databases, i.e. it makes advanced tools accessible to a wider public by means of an easy to use GUI.

As a rule, RM communicates with a cloud. This is a very undesired feature when it comes to professional use and adherence to corporate standards. I have tried several times to discuss the issue with RM's customer support, but there seems to be no good understanding how to savely use RM locally (i.e. without an internet connection) in a way that is complient with corporate standards regarding IT/data security. This is very sad since it limits the professional use of RM to a design phase, while for production you still have to do the implementation in Python or the like.

Must try!

Any sort of data analysis using supervised and unsupervised learning algorithms, typically on numerical data (no text mining).

The design of the tool and associated tutorials that guide the first time users and learners to start analytics and machine learning first by scratching the surface and then slowly drill deeper. The design that centers around use of operatives, categorized into different sets like Data Cleansing to Predictive Analytics provide a visual guide to learners about the steps necessary in building models.

Nothing much. RapidMiner need to expand and include more operatives like say statistical tests to test stability or normality. Often statistical tests form important part in model building.

Generally in survey analysis. RapidMiner helps in surveys in classification of respondents based on profiles and specificity in responses.

As a person who has taught myself analytics (and continues to do so), what I appreciate most about RapidMiner Studio is the embedded tutorials.

In 7.xxx, I did not like the UI so much. Now, in version 8.xxx, the UI is a bit nicer. Still, the windows such as a parameters and operators can be hard to scale and position on smaller screens (such as my main notebook) though they display well on external monitors.

Give it a try, especially if you are a beginner!

Right now getting a foothold in predictive analytics--I am gaining skills that I plan to apply to an upcoming e-commerce website project.

RapidMiner Studio is free to use for educational purpose. This is the main purpose I use this softwre so it fits quite good. The licences are straightforward to obtain.

On the technical side, it is quite easy to use for the students with little technical background like my students but it becomes a poweful tool even for the programmers.

Not much, perhaps some more features to use it in batch mode, obtaining and assessing the results automatically. When grading 80 solutions for a problem, it becomes a tedious work to open each solution sent by the students and run it. I have used the batch mode with some success.

The students are able to implement their solutions to Data Mining problems presented during the course.

User-Friendly GUI-based process modeling

Easy parameter setting

Auto Model

Data Quality

User technical support sites, such as Resource site, Youtobe, chatbot, Community, etc.. need to be more organized and user-friendly.

extensions for Korean characters, such as word2vec, text mining

Process Model Knowledge base needs to be more organized and updated

Unclear response from Chatbot customer services

Week of cutting edge Deep Learning functions and blenders.

Knowledge bases needs to be improved

Extension partner's sit

Definitely I recommend. but it needs to support Korean Character set for text mining and word2vec.

I'd be a contributor for Korean market localization because of Korean NLP and POS tagger experience.

Vod(Video on Demand) recommendation project for Korea Telecom.

AI AML for Korea FIU

AI semantic regulation for Korean Financial Supervisory Services

It is a visual programming based software that is easy to learn, easy to use, intuitively help users to understand and focus on the process and results rather than thinking on coding. Of course we can't run away from tuning the parameter but it just make your task simpler. Furthermore, the features of RapidMiner can be significantly enhanced with other add-ons, many of which are also available for free.

It is not intelligent enough to detect unwanted or hidden character in the dataset. Sometimes RapidMIner can't detect that the input is a date but recognize it as nominal. I wish in the near future, RapidMiner will add additional intelligent sensing especially when the data comes from legacy database systems, or even Excel. I hope most of the powerful add-ons by third party companies will make it free for education purposes.

I have been considering other tool for teaching, but one of the factor I use RapidMiner in teaching is their program for education. You don't feel like you're begging for 'free lunches'. The support staff are friendly and their response time to our needs are pretty awesome. That was my experience. Try it. Nothing to lose.

I use Rapidminer to teach undergrad and postgraduate students on data mining, decision support and many kind of analytics such as descriptive, diagnostics, predictive and semantics. I think my students can grab the concept of analytics faster than other tools that I have been using for a while.

The software is easy to use. The user does not have to code, but just have to drag-and-drop the operator. It also has many extensions regarding text processing and machine learning process.

The explanations for some parameters are not clear enough. For people who do not have any knowledge about the algorithm, they will be confused whether they should modify the parameters or not.

I use RapidMiner to do my Master thesis about building methods for automated grading. Since I don't have time to learn to code in any programming language, I found RapidMiner is easy to use and requires little time to learn about it. The community forum is also helpful to find some issue and the solution to solve it.

The user interface makes it very easy to tell the machine what to do. So even people without programming experience can set up data mining tasks. There is much support like videos, tutorials, documentation and a huge user community. With just a bit of instruction at the beginning you can gain fast progress.

Almost nothing. Very special problems could be solved more directly in Python.

It's easy to start. So just try it.

We are doing natural language processing. RapidMiner helped us to set up and test a variety of text mining approaches.

- Intuitive interface

- Data input very intuitive and easy

- Most machine learning algoritms are there, don't need to look for others

- Good documentation with tons of examples on product capabilities and features

- Responsive and active community always willing to help

- Poor graphic plotting possibilities, very little option to configure plots

- Once you leave the platform, all graphs you generated are gone, have to recreate them.

If you wan't to quickly learn how to machine learning works, this is the tool for you, you don't need any deep knowledge in the field. My recommendation is to start with Gradient Boosted Algoritms, since they are very robust and overfitting prone.

Mostly agriculture related problems, like creating models of plant nutrient content via spectroscopic techniques, etc.

I used it as a recommender system and I found it awesome

No dislikes they only get better I strongly believe when more clients get to know about them many will leave r because when the work is done the client needs to manage the process on his own. Rapid miner gives the client a soft landing when consultants are done with their work. He can simply take over and run algorithms on his own without no knowing how to code

The latest version proves to me that it can be used to predict machine breakdowns awesome tool looking forward

Rapidminer is one of the best data science software, I appreciate how rapidminer works on different data, it covers almost all of the algorithms you need to produce an output you want. Rapidminer had help me in analyzing different data set for my project. Lastly, I like rapidminer because even though a person who doesn't know how to program R or other languages, he/she can use rapidminer because it has understandable and interactive graphical user interface. The software is not free, yes, but despite being able to use its commercial license, there are a lot of things you can also do.

Rapidminer supports almost all of the needs of the user, but despite all of that, on my part as to having a laptop (4GB ram) as an own device, it is kinda hard to run some of the processes because it eats almost all of my ram but it depends on the device. On my desktop (8GB ram), this software runs without any bugs or error.

There are no business problems as of now.

Easy and fast learning curve.Official online Video Tutorials are very good. I'm new to Data Analytics (11 Months).In a very short space of time, The default screen is very well designed. The "Wisdom of the Crowds" section is very useful. There were plenty of times I was able to select the operator I needed from this section, This saved time. I could do Predictive Modelling and Sentiment Analysis. With reference to Sentiment Analysis, I wanted to compare two data sets. I was able to create a process, apply it to my first dataset and save the process. I was then able to reuse the same process on the second dataset. There was no need to rebuild the process. This saved me a considerable amount of time.

There is no coding involved. I could Fine Tune the process and Parameters.This allowed me to concentrate on producing better reports.

Rapidminer has the ability to produce fantastic charts and graphs. There is also the option to export results to Tableau.

I found it difficult to find an Official online Video to link up MYSQL database to Rapidminer.

This is probably more of a technical issue than Rapidminer.

Predictive Modeling, I used different models Decision models for example Decision Tree, Random Forest, Naive Bayes and K-Nearest Neighbour.

Sentiment Analysis. With reference to Sentiment Analysis, Search Twitter operator.

Third Party Extensions Analyse Sentiment Operator (Aylien.com) and Extract Gender Operator(Namsor.com)