Ease of use. DTREG is a robust application that is installed easily on any Windows system. DTREG reads Comma Separated Value (CSV) data files that are easily created from almost any data source. Once you create your data file, just feed it into DTREG, and let DTREG do all of the work of creating a decision tree, Support Vector Machine, K-Means clustering, Linear Discriminant Function, Linear Regression or Logistic Regression model. Even complex analyses can be set up in minutes.

Classification and Regression Trees. DTREG can build Classification Trees where the target variable being predicted is categorical and Regression Trees where the target variable is continuous like income or sales volume.

OUR CLIENTS

Our research focuses on prediction of maximal oxygen uptake, performance measures of multiprocessor architectures and upper body power of cross-country skiers using different machine learning methods. With no doubt, DTREG is a great tool that provides us with these methods along with a user-friendly interface. I definitely recommend DTREG to people from different fields who work in the area of applications of machine learning methods.

M. Fatih Akay, associate professor, Computer Engineering Department, Cukurova University, Turkey

Jan 07, 2015

I've found DTREG very useful to explore ecological data sets in order to reduce the number of variables down to a more manageable number and to isolate relationships among them. DTREG's graphical interface allows you to drill deeper down into statistical output without the additional programming and customization that some open source software packages require. DTREG gets you their faster.

J. Anthony Stallins, University of Kentucky

Jan 07, 2015

We have been using DTREG software for more than 5 years for predictive modeling of many types of data sets, including medical records and measurement data obtained from the manufacturing machines. The classification and regression models created by DTREG can be easily adjusted, and their quality can be neatly evaluated. DTREG is a well-designed and very flexible tool. It is probably the easiest software to use, in comparison with several other programs for data mining.

Jacek Kluska, Professor, Computer and Control Engineering, Rzeszow University of Technology, Poland

Jan 07, 2015

The Biomedical Engineering and Telemedicine Group of the University of Cádiz (Spain) uses DTREG software in both teaching and researching activities. DTREG tools for data pre-processing, classification, regression, clustering and scoring are widely applied to a number of biomedical contexts.It is an intuitive tool, easy to use, that allows implementing in a fast and effective manner multitude of experiments using data mining techniques.

Daniel Morillo, University of Cádiz, Spain

Jan 07, 2015

I have taught Biostatistics courses to graduate students and medical residents for more than two decades, specializing in time series analyses and artificial intelligence methods. DTREG is extremely useful not only for its user-friendliness but also its capability in a wide range of applications in higher education and public health. I think that you have a unique software with compelling modeling approaches.

C.K. Chen, EdD., Director of Institutional Research, Meharry Medical College

Jan 07, 2015

I have been using DTREG for 2 years to classify neuroimaging data of neuropsychiatric disorders and normal controls. The software is well suited for these purposes and the support is helpful.

Tetsuya Iidaka, M.D., Nagoya University

Jan 07, 2015

DTREG is very useful for geophysists who make studies using the well log data and seismic data in their research. I am very happy for the results obtained from this software. Thank you very much for DTREG software.

Rafik Baouche, University of Boumerdes, Algeria

Jan 07, 2015

As a professional energy trader, I have found DTREG to be exceedingly useful in helping me organize important data and to make predictions about the future. I have used DTREG for projects over the past two years from the creation of regional supply and demand curves for wholesale power by ISO to predicting future natural gas futures prices. I have consistently achieved excellent results. Hats off of DTREG!

Kevin Capone, Head of Energy Trading, Sierentz North America LLC

Jan 07, 2015

DTREG is a great affordable tool for diverse business applications of data mining/machine learning: product sales analysis and forecasting, product sales drivers identification, response curve development, dissimilar CRM problems, etc. DTREG includes a variety of methods that can be found in expensive predictive modeling software, but DTREG also provides the capability that are not included in those software. We have been using DTREG for 5 years, and we recommend DTREG without reservation.

Pavel Brusilovskiy, Ph.D., Senior Specialist, Quantitative Sciences, Merck & Co.

Jan 07, 2015

I have used DTREG since 2013 for making papers about forecasting processes needed in the management in the Civil Engineering, where important decisions should be made like: forecasting time and cost of construction, bidding price for designing and construction. DTREG is predictive modeling software which is very easy to use, and it provides the best state-of-the-art modeling methods including neural networks and SVM.

Silvana Petrusheva, Associate Professor, Civil Engineering, Skopje, Macedonia

Jan 07, 2015

DTREG contains most of the machine learning algorithms for classification. The best part about DTREG is the ease of use. One needs minimal training to start running it (unlike open source softwares where there is a steep learning curve) and can get models off the ground in rapid time. Deployment of developed models is also easy as DTREG exports the model code with just a click of a button.

Shivakumar Raman, Northwestern University

Jan 07, 2015

I have been using DTREG for analyzing the controlling factors for the recovery of oil from various conditions. It really helped my understanding of the key controlling recovery factors based on large number of variables and samples in a very efficient way. Phil has been really a great help in my research using DTREG.

Shengyu Wu, Ph.D., Director of Geosciences, C&C Reservoirs

Jan 07, 2015

I have used DTREG for more than five years to do the data analysis and build predictive models for my business projects at Mead Johnson. DTREG includes several unique data mining algorithms, such as TreeBoost, Gene Expression Program (or Symbolic Regression), Decision Tree Forest, and so on. I find it to be easy to use, efficient, and comprehensive. I am happy to recommend DTREG.

K. T. Chen, Manger, Marketing Analytics, Mead Johnson Nutrition

Jan 07, 2015

"I have used DTREG GMDH module from December 2013 for data mining for soil science for my PhD research. It is impressive that the software is easily for use with controllable, tractable and easy understandable output report. And also, the software can score data using established models. Hence, DTREG is very useful and helpful tool for the large data mining."

Zou Lei, Nanyang Technological University, Singapore

Nov 21, 2014

"I have been using DTREG for analyzing the controlling factors for the recovery of oil from various conditions. It really helped my understanding of the key controlling recovery factors based on large number of variables and samples in a very efficient way. Phil has been really a great help in my research using DTREG.

S. Wu, Director of Geosciences, C&C Reservoirs

Nov 20, 2014

"DTREG is one of the best and most comprehensive machine learning software's on the market. The user-interface makes it extremely easy to analyze and compare multiple machine learning algorithms on a given set of data. The code-generation option makes it practical and easy to integrate a predictive model with your existing application. The documentation and technical support provided is top-notch!"

Tejas Mehta, Sr. Algorithm Engineer, Rapiscan Systems

Nov 20, 2014

SOLUTIONS

Predict Political Party Example

Type of tree: Time series analysis
Task: Perform time series analysis to forecast airline passenger loads. A Gene Expression Programming model was built for a time series. This model was then used to forecast future passenger loads....
Keep Reading...

Boston Housing Cost Example

Type of tree: Regression
Task: Predict house values based on locale This is a regression tree example to predict the value of houses in various areas around Boston based on characteristics of the locale such as proximity to the Charles River and major highways, socioeconomic status, air pollution and other factors. The target variable is house value....
Keep Reading...

Landing Control Example

Type of tree: Classification
Task: Determine if it is better to use automatic or manual landing control This is a classification problem to decide whether it is better to use manual or automatic (autopilot) control when landing the space shuttle. The target variable has two categories, Automatic and Manual. The predictor variables include wind direction and velocity and visibility....
Keep Reading...

Liver Disorder Example

Type of tree: Classification
Task: Predict normal or abnormal liver condition This is a dataset from England that generates a classification tree to predict liver disorders. The target variable is liver condition (Normal or Abnormal). The predictor variables are various blood chemical measurements along with the number of alcoholic drinks consumed per day....
Keep Reading...

Iris Classification Example

Type of tree: Classification
Task: Classify species of iris based on plant measurements This is a classification problem dating back to 1936. Its originator, R. A. Fisher, developed the problem to test clustering analysis and other types of classification programs prior to the development of computerized decision tree generation programs. The dataset is small consisting of 150 records. T...
Keep Reading...

DTREG Benchmarks of Predictive Model Methods

Benchmarks of Predictive Model Methods The following table shows the results for various types of predictive models applied to a large number of benchmarks. Note that different types of models work best for different types of data.

All of these benchmarks are classification problems (i.e., the target variable is categorical). Some types of models work better with regression problems where t...

Keep Reading...