Salford Systems specializes in state-of-the-art machine learning technology designed to assist data scientists in all aspects of predictive model development. Salford's tools are known for their ease of use, capability of working with large volumes of data, high-speed model development, robustness and reliability and consistent delivery of ultra-accurate models. Salford's modeling automation tools guide novice data scientists through the complex process of model development and help expert data scientists develop world-class predictive models. Salford software has been used in over 3,500 organizations worldwide (financial services, insurance, transportation, retail, healthcare, science, and high tech) and has been deployed in online targeted marketing, credit risk scoring, financial fraud detection, insurance risk management, logistics, bio-medical research, manufacturing quality control, and dozens of other fields.
Reliable, easy to use, and easy to understand data mining tools are increasingly in demand as stores of customer and business information grow in corporate data warehouses and data marts. Corporations that leverage these data mining tools to develop better predictive models and to better understand their customer base are able to make more profitable long term business decisions. Salford Systems is spearheading data mining tool development by maintaining an active R&D program, staffed by Ph.Ds trained at Harvard, MIT, and UC Berkeley, and leveraging existing ties to leading universities.
Salford System's flagship data mining software is an affordable and robust multi-platform classification and regression tree package. CART automatically generates a wide variety of highly accurate data mining analyses. It is the only decision tree tool based on the proven methodology of the original CART code, which was developed by world-renowned Stanford University and University of California at Berkeley statisticians. Designed for both non-technical and technical business users, CART can quickly reveal important data relationships that could remain hidden using other data mining tools. In addition, it offers flexibility and advanced capabilities for professional, production level applications.
Jerome Friedman's MARS (Multivariate Adaptive Regression Splines) is stepwise regression done right for the first time. MARS does variable selection, variable transformation, interaction detection, and self-testing to prevent overfitting, all automatically. There is only one trademarked MARS and it is available exclusively from Salford Systems.
Jerome Friedman's latest data mining tool is based on boosted decision trees. TreeNet® is an astonishingly accurate model builder and function approximation system that also serves as a powerful initial data exploration tool. Use TreeNet® to extract the most important relationships in your data and calibrate how predictable the outcomes are. Then either use the TreeNet® model directly or incorporate the results in CART, MARS, or conventional statistical models.
Leo Breiman's latest data mining technology is based on learning ensembles of CART trees. By judiciously injecting randomness into the tree-building process and then combining hundreds of these trees, RF is able to deliver high performance predictive models and a variety of novel exploratory data analysis results. RF also incorporates new metric-free CLUSTER analyses that automatically select the variables used to define each cluster, with potentially different variables defining each cluster.
The SPM Salford Predictive Modeler® Software Suite
The SPM Salford Predictive Modeler software suite is comprosed of Salford Systems' core data mining engines (CART, MARS, TreeNet, RandomForests) as well as additional features geared to improve predictive models. It is a highly accurate and ultra-fast platform for developing predictive, descriptive, and analytical models from databases of any size, complexity, or organization.
Salford Systems offers corporations and management consulting companies a variety of analytical and strategic consulting services. Salford teams business consultants and technical Ph.D.s with scientific programmers to find innovative solutions for complex modeling and data analysis problems. Areas of specialization include database mining, segmentation, targeting and choice modeling. Salford Systems maintains a rapid response data mining center equipped with six high speed servers and massive storage capacity. Demonstration cost-effective projects and proof of concept studies can be planned and executed in as little as four weeks, and assessments of the value of large scale data mining projects can be generated quickly. Salford Systems conducts large scale data mining projects from initial conceptualization to the final installation of productivity software.
To bolster its services, Salford Systems offers an ongoing series of database mining training seminars for CART and other innovative tools. Companies can receive on-site training or send representatives to seminars presented periodically in most major U.S. cities.
The FounderDan Steinberg, Ph.D.President and Founder
Dan Steinberg, President and CEO of Salford Systems, founded the company in 1983 just after receiving his Ph.D. in Economics at Harvard. He also served as a Member of Technical Staff at AT&T Bell Laboratories and Assistant Professor of Economics at the University of California, San Diego, and has participated in dozens of consulting projects for Fortune 100 clients. He has been honored by the SAS User's Group International (SUGI) and led the modeling teams that won the KDDCup 2000 and the 2002 Duke/Teradata Churn modeling competition. Dr. Steinberg has published articles in statistics, econometrics, computer science, and marketing journals, and has been a featured data mining issues speaker for the American Marketing Association, American Statistical Association, the Direct Marketing Association and the Casualty Actuarial Society.
How and Why Salford Systems got started
Dan Steinberg was always interested in analytic methodology and in software to make the methodology easy to use, from his days as a grad student in Econometrics at Harvard. The company started with statistical products for the IBM mainframe in 1983 and shifted development efforts to the PC in 1985.
Two distinguished professors and a network of colleagues have been key to our innovation strategy. We began by building on the genius of Leo Breiman (Stats UC Berkeley) and Jerome H. Friedman (Stats, Stanford) and combined their groundbreaking technological innovations with our own experience as practitioners of data analysis and predictive modeling. Breiman contributed new ideas up until a year before his death in 2005, and Friedman continues to provide us with innovations. We regularly tackle challenging problems for major corporate clients, and are thus stimulated to devise new solutions along with improvements to our core technologies.