New “Office add-in” offers easy to use forecasting, machine learning, data mining and text mining in a web browser, using a point-and-click interface familiar to millions of business users.

INCLINE VILLAGE, NV – June 24, 2019 – Frontline Systems, developer of the Solver in desktop Microsoft Excel®, has released Analytic Solver® Data Mining V2019, a SaaS (software as a service) cloud and desktop platform that offers Excel-savvy business analysts point-and-click tools to create forecasting, data science, machine learning, and text mining models themselves, without help from data scientists or programmers.

Available now in Microsoft AppSource, Analytic Solver Data Mining is the only currently-available data mining and machine learning software that works in a browser with Excel Online, as well as in desktop Excel for Windows and Macintosh.  A companion ‘Windows desktop-only’ version is also available for download from www.solver.com, for users without an Office 365 subscription.

Analytic Solver V2019 has been rewritten from the ground up using JavaScript, REST API, Azure and Office 365 technologies, but it inherits the full power and ease of use of Frontline Systems’ analytics software from over 25 years’ experience supporting business analysts using these methods in practice.

“For millions of business users, Excel is a powerful, yet easier to use tool than R or Python for data science and machine learning,” said Daniel Fylstra, Frontline’s President and CEO, “and now we’ve brought this power and ease of use to Excel Online.”

User Interface Familiar to Business Users

Analytic Solver V2019 offers a user interface with a Ribbon and drop-down menu choices for individual forecasting and machine learning functions, a Task Pane for building data science workflows, and the full range of Excel features for model building, whether it’s used in a browser, on Windows, or on Macintosh.  All the elements of data science models are saved in the user’s Excel workbook, which can be transferred freely between cloud and desktop.  Unlike other predictive analytics tools, models and their results are ‘fully transparent’ as Excel spreadsheets, readily understood at all levels of management.

Using Data and Models from Cloud and Desktop Sources

Loading data sets, query-based summaries, and representative random samples of data from SQL Azure databases, and Apache Spark Big Data clusters is a simple point-and-click operation in Analytic Solver V2019.  Users who are learning – via Frontline’s own online courses at https://Solver.Academy, or the hundreds of university courses that use Frontline’s analytics software – can easily choose from a variety of example models and datasets, including Big Data examples using Frontline’s always-on Spark cluster.

Forecasting, Data Mining and Text Mining

Users can explore data using a Chart Wizard, feature selection, principal components, and k-means or hierarchical clustering; create exponential smoothing and ARIMA models for forecasting; perform text mining with latent semantic analysis; apply multiple regression, k-nearest neighbors, and ensembles of regression trees and neural networks for prediction, or discriminant analysis, logistic regression, naïve Bayes, k-nearest neighbors, and ensembles of classification trees and neural nets for classification problems; and use association rules for recommendation applications.

Easy Model Deployment to Web Applications

Deploying a data science model in a production application was once a very difficult process, but Analytic Solver Data Mining makes it easy.  Optimally fitted forecasting and trained machine learning models are created in PMML (the industry-standard Predictive Modeling Markup Language), usable in a variety of applications including the CRAN Pmml package for R.

Even better, predictive models plus decision logic translating predictions into actions – using simple Excel formulas – can be deployed “point and click” using a Create App menu option, yielding a web page that runs the model using JavaScript and Frontline’s RASON® Analytics API. RASON models also work readily with C/C++, C#, Java, R and Python, and are flexible enough to run on desktop PCs, on-premise servers, or on websites or in mobile apps, using REST API calls to Frontline’s cloud service.

Frontline Systems Inc. (www.solver.com) is the alternative to analytics complexity, helping business analysts and managers gain insights and make better decisions for an uncertain future, without the cost, delays and risk of ‘big vendor’ tools.  Its products integrate forecasting and data mining for “predictive analytics,” Monte Carlo simulation and risk analysis, and conventional and stochastic optimization for “prescriptive analytics.”  Founded in 1987, Frontline is based in Incline Village, Nevada (775-831-0300).

Excel® is a registered trademark of Microsoft Corp. Analytic Solver® and RASON® are registered trademarks of Frontline Systems, Inc.