Findability Sciences launches Muti-algorithmic Forecasting capabilities in its Platform

WOBURN, Mass. - March 4, 2020 - PRLog -- A leading Data Sciences and AI company, Findability Sciences, announced today launch of a new capability for forecasting through multiple algorithms in its industry leading offering, Findability Platform®. Eleven forecasting algorithms have been integrated with the Findability Platform®, to meet a broad range of business requirements, which includes integration with usage based, paid services from firms like Amazon Web Services (AWS). The locally hosted forecasting algorithms include all-time favorites such as ARIMA (Auto-regressive Integrated Moving Average), REG.ARIMA (Dynamic Harmonic Regression ARIMA) and ETS (Error-Trend-Season), as well as NNETAR (Neural Networks Time Series Forecast), Prophet, BATS and TBATS etc.

While making this announcement, Suresh Shakkarwar, company's head of product development, said "This was a long due requirement from our customers and partners. In the rapidly changing environment of Data Sciences and AI, users prefer all the options under one umbrella to compare results from various options and chose the one best suited for their requirements"

With a wide range of algorithm choices, Findability Platform allows 'ensemble' or combining of algorithms to develop perfect forecasting models. For example, NNETAR may be deployed in combination with ARIMA to generate better forecasts. Therefore, depending on data and business problems, users can experiment with multiple algorithms and multiple sets of algorithms. Users can also test different combinations of input variables and fine-tune the hyper-parameters of forecast models to optimize results.

"This capacity, combined with our recently formed joint venture with Qlytics, offers tremendous opportunities for Citizen Data Scientists and seasoned Data Scientists at our customer organizations, to deploy supervised learning algorithms in a collaborative environment" said Anand Mahurkar, CEO of Findability Sciences. The enhanced forecasting capabilities of Findability Platform® are expected to find widespread applications in automotive, manufacturing, retail, healthcare, logistics, commodity trading and hospitality sectors. Estimating product demand, predicting weather, inventory planning, estimating web traffic, forecasting room occupancy, commodity and stock price prediction, etc. are some examples of the business decisions which can be facilitated by the Platform.

About Findability Sciences

Founded in 2010, Findability Sciences is an AI, Big Data, Cognitive Computing firm, headquartered in Woburn, MA, with subsidiaries in Canada, India and Japan. Our flagship product, the Findability Platform^® is a supervised-learning advanced analytics product powered by proprietary technologies to deliver turnkey business solutions. A portfolio company of SoftBank Corp, Japan, we are part of the IBM Watson Partner Ecosystem.