STATISTICA Data Miner: Uncover Hidden Trends, Explain Known Patterns, and Predict the Future

StatSoft, Inc. recently announced the release of

From querying databases and drilling down to generating final reports and graphs, STATISTICA Data Miner offers ease of use without sacrificing power or comprehensiveness. STATISTICA Data Miner features the largest selection of algorithms on the market for classification, prediction, clustering, and modeling as well as an intuitive icon-based interface.

The data mining solutions in STATISTICA Data Miner are driven by powerful procedures from five modules: General Slicer/Dicer with OLAP, General Classifier, General Modeler/Multivariate Explorer, General Forecaster/General Neural Networks Explorer. The complete functionality of STATISTICA, which has received the highest rating in every comparative review of statistics software in which it was featured since its first release in 1993, is also available for data mining. All analyses and methods are highly optimized for speed and efficiency and can be connected to quickly compare different approaches.

Users can choose from hundreds of graph types to visualize their data after cleaning, slicing, and drilling down and can even publish these results via the Web.

STATISTICA Data Miner can work in client-server architecture using optional WebSTATISTICA Server applications. WebSTATISTICA Server applications offer the ultimate enterprise system functionality, including a browser-based user interface, options to offload time consuming tasks (e.g., complex queries or summaries of large data sets) to the servers, distributed processing (scalability to multiple data processing servers), and options to manage projects over the Web and work collaboratively across the corridor or across continents.

TECHNICAL NOTE
STATISTICA Data Miner is the only data mining product on the market that offers the ease-of-use of a “point and click” user interface and the flexibility of programmability and customizability of the system. STATISTICA Visual Basic allows you to access the complete functionality of all STATISTICA products via more than 10,000 functions comprising one of the largest development environments available. This open architecture enables users to access the full power of the analytical routines, hundreds of analytic and descriptive graphs, and specialized routines for data mining in STATISTICA Data Miner and enables users to customize the system to include third-party or in-house proprietary algorithms and methods.

Programming STATISTICA Data Miner in STATISTICA Visual Basic is extremely easy, not only because it is industry standard and comes with a programmer-friendly development environment, but also because the “programming” can be done by recording logs of interactive data analyses. However, for customers who need a complete, deployed, and ready-to-use solution designed to solve a specific types of problem, STATISTICA Data Miner is offered with optional deployment and on-site training services.

WebSTATISTICA Server Applications - provide functionality for full Internet enablement to the comprehensive and powerful data analysis tools in STATISTICA, including the ability to run STATISTICA from a Web browser, and enable users to easily and quickly access data and powerful analytical tools from virtually any computer in the world as long as it is connected to the Web. WebSTATISTICA Server applications can be either purchased as stand-alone products accessible to users who are connected to the Internet or as an extension of an existing STATISTICA product installation.

Back to Top

STATISTICA Six Sigma Green and Brown Belt Training Available
StatSoft, Inc. proudly announces the addition of Six Sigma Green and Brown Belt training to its comprehensive array of training course topics. This innovative Six Sigma training program combines traditional classroom-based teaching with multimedia instruction to provide participants and employers with an effective, convenient, and cost-efficient introduction to the Six Sigma methodology and tools.

Sign up today for StatSoft's Six Sigma training and Achieve your company's Six Sigma Goal:

• Get a complete overview of Six Sigma benefits
• Obtain knowledge of Six Sigma QC/QM tools
• Apply tools and techniques to real-life examples
• Learn how to facilitate improvement teams

Lead your organization to a higher level of quality today! For more details or to register, click here.

In addition to a broad range of products for data analysis, data mining and quality control applications, StatSoft offers a variety of training and consulting services worldwide. Click here for details.

Back to Top

First Review of STATISTICA 6 - "THE FUTURE IS HERE...Superior Performance, Enhanced Handling, and No Glitches"
STATISTICA 6 (Beta) recently received a very warm welcome from reviewer Felix Grant in the review "The future is here...at last" in the July/August issue of Scientific Computing World. Grant begins by reviewing the most obvious change from previous version, the interface, and describes the STATISTICA 6 interface as "extensively customizable to reflect your own requirements." He continues, "This ability to shape a working environment in the interests of maximum personal or team usefulness has, of course, always been one of STATISTICA's strong points; but the means now reflect current Windows trends."

He goes on to compare the functionality in STATISTICA favorably against that of one of the most popular software products in the world, "Menus adapt automatically to application frame context, in a way that is better managed and more flexible than Microsoft's, and customization can be made to reflect this. So far, I have not yet hit the limits of this customization potential."

In the remainder of the review Grant praises many other areas of the program, including the new integrated STATISTICA Visual Basic - "well cool", the workbook system - "a definite advance," and many others... "Graphics benefit from an impressive range of incremental improvements with new types, more visual and intuitive access selection, enhanced brushing and zooming, style management, and interactive sectioning of 3D constructs. There are new statistical procedures in modelling and data mining, and improvements in a swathe of others. The whole package, never slow on its feet, has gained noticeably in speed and responsiveness."

After running STATISTICA alongside competing products Grant concludes, "I have become a fan (of STATISTICA), encountering only superior performance, enhanced handling and no glitches."

StatSoft and STATISTICA receive top marks in a survey conducted by SCIENCE
A new survey conducted by SCIENCE (the publication of the American Association for Advancement of Science; circulation: over 150,000) of a randomly selected sample of its readers has provided data on how StatSoft, Inc. is perceived by the community of scientists and administrators of science. The following are unedited statements about StatSoft, Inc. and its products volunteered anonymously by the respondents. The data for the surveys, sponsored by SCIENCE magazine, were collected by the Harvey Research Organization, Fairport, New York. The following selection of excerpts has been accepted by SCIENCE for use in this material (reprinted by permission of SCIENCE and Harvey Research Organization):

"I have used StatSoft [STATISTICA] before(...). You can't beat them."
-- Professor of Biology, University

"I know a lot of people that use it (...). The [STATISTICA] graphics are awesome."
-- Research Assistant, University

"We use this in many of our departments. The university appreciates its versatility."
-- Professor of Biology, University

"They make excellent statistical software."
-- Genetic Toxicologist, Industry

"(...) They have a great series of software packages."
-- Associate Professor, University

"(...) I think that StatSoft has a good thing going with this product. You can do a lot with it."
-- Associate Professor, University

"(...) I use their statistical package. I need to get another program from them. They have good quality."
-- Professor, Medical School

"I do know this company. I have a good image of them."
-- President, Industry

"I am familiar with this company. I have a good opinion of them."
-- Program Leader, Industry

"I think that StatSoft is excellent."
-- Vice President of R&D, Industry

"I like their statistics package because you can use it in all different fields. We have used it to make some of our displays. (...) I have always been impressed by them."
-- Botany Curator, Museum

"They have some excellent software. Some of it is very useful. I think the company is on the cutting-edge.""
-- Microbiologist, Government

"I have a good image of this company."
-- Asst. Professor, University

"(...) I am very impressed with them."
-- Sr. Research Associate, Laboratory

A Comparison of Statistical Software Packages Finds STATISTICA "Flexible and Widely-used" Among Both Experts and Beginners
The July 9th issue of The Scientist featured a brief overview of some of the most commonly used statistical software tools on the market, and STATISTICA's description clearly stands out among the descriptions of its peers.

In his introduction to the comprehensive statistical tools available, Author Paul Wolf notes that "Several large packages are geared toward the statistical expert. These are often more flexible, and some can handle larger and more complex data than their smaller brethren can. Familiarity (and effective instruction) can make even the most cumbersome software user-friendly, but the general trend is toward ease-of-use across the spectrum."

However, later in the same section, he introduces STATISTICA by saying "The difference between simple and relatively inflexible packages and complex and versatile ones is not necessarily clear cut." Wolf describes STATISTICA as "a very flexible and widely-used application that could appeal to both experts and beginners" and goes on to say, "the entire approach to the software is scalable, and quality of output is excellent".

Click here to view other STATISTICA awards, comments from users, and a complete summary of STATISTICA's unmatched record of reviews.

Back to Top

INSPEX 2001
Birmingham, UK - National Exhibition Centre
October 30 - November 1, 2001
Stand 8660

ICDM 2001 (IEEE International Conference on Data Mining)
San Jose, California, USA - Doubletree Hotel
November 29 - December 2, 2001

CHIMIOMETRIE 2001
Paris, France
December 4 - 5, 2001

National Forum on Quality Improvement in Health Care
Orlando, Florida, USA - Orlando World Center, Marriott
December 10-11, 2001
Booth 205

Click here to view the complete list of exhibits StatSoft will attend in 2001-2002.

StatSoft offers both introductory and advanced training courses in major cities in the United States and overseas. StatSoft's training classes offer:

  New Features in STATISTICA 6 and Intro to Visual Basic - This one-day course is designed to introduce users of version 5 to the wide array of new features that allow STATISTICA 6 to break new barriers in data analysis, data mining, and QC applications.

  Introduction to Visual Basic and STATISTICA Visual Basic - This one-day course is designed to teach STATISTICA users how to use automatically recorded Visual Basic macros to effectively automate and customize interactive procedures.

  Visual Basic Applications in STATISTICA - This one-day course, designed for the user familiar with STATISTICA and with the fundamentals of Visual Basic programming, will introduce the STATISTICA Object Model, which captures the functionality of the interactive program in over 10,000 functions accessible through Visual Basic.

A very good function in VGLM
If you want to get the predicted value of your response variable of a set of known predictor levels according to the model, you can get the predicted values very easily. If you wish to get a batch of predicted values, you can do this as well: First in your data spreadsheet, add some cases, enter the values of your predictors, but leave the response values empty (missing). Proceed analysis as usual; after you get your result window, go to residual tab, choose Predicted option (the default option is Analysis), click Predicted and residuals button; you will get a scrollsheet with the predicted values that you are looking for.

Binomial Distribution
Question: I have conducted analyses based on a binomial distribution using a logit link function, and all of the parameter estimates I get appear to be opposite of what they should be.
Answer: In VGLZ, with the binomial response function, the program takes the first code that it finds in your data sheet and gives it a value of 1 and the second code becomes 0. For example, if the text for the first code is "No" and you have assigned it a value of "0", STATISTICA will flip the coding and recode the "No" as a "1". The way to keep your original coding is to not allow the program to select the codes, but after selecting the variables in your analysis, select the Response Codes box and enter 1, 0 in the box. This way, the program will see the code 1 first and assign it a value of 1.

---

Please note: StatSoft, Inc. will never share the email addresses of its subscribers with any company or organization and will never make them public. The list of our subscribers is treated as privileged information and is well protected. Also, you may unsubscribe from the StatSoft News letter at any time by creating an e-mail with "Unsubscribe" in the subject line and sending it to

 Pacific
Suite 1, 46-48 Howard Street
North Melbourne VIC 3051
Australia
Phone: +61 3 9348 9422
Fax: +61 3 9348 9420
e-mail: info@statsoft.com.au

©Copyright StatSoft, Inc., 1984-2006.
StatSoft, StatSoft logo, STATISTICA, Enterprise/QC, Enterprise, Data Miner, SEPATH and GTrees are trademarks of StatSoft, Inc.