Our professional consultants have expertise from over 10 years of experience building data science models, predictive algorithms, and big data solutions. So we are well equipped to solve your major data challenges. We are pioneers in the industry, doing “data science” before it even became a popular term. We work with many fortune 500 companies, as well as smaller firms. We know how to make a big difference in your business. Read some of our success stories below.
Our Bootcamp students are skilled and eager to help you with an immediate project need. We admit only the highest quality applicants, most with PhDs and Masters degree from all areas of science and math. They come to our Bootcamp to learn how to apply their already mature skills to this new and exciting field of data science. For no fee, you can work with our students to create a model that will visualize your data, explore insights buried in your data, and create a predictive model to help you make better business decisions.
Problem: Holiday season dramatically affects car sales. Being able to predict seasonal car sales to support inventory and staff management is critical for all manufacturers.
Solution: SupStat was able to build and test multiple ARIMA time series models to reliably predict total car sales with contributions from different factors. The results were then integrated into the predictive computation within the client’s web framework.
Problem: Telecommunications billing data has exploded in recent years due to the rapid development of mobile technology and changes in usage patterns. Storage and retrieval on this scale poses challenges to the speed and efficiency of data analysis and management.
Solution: SupStat used a Hadoop platform on x86 clusters to provide a 30-fold increase in query performance over RISC platforms. This new system can handle 30 Terabytes of monthly users’ billing data. The SupStat models help to identify individuals who check their billing information and predict who will check it and then prepare the billing summary
Problem: Major transit authority recognized that many passengers were being left on the platform unable to get into trains that were too crowded.
Solution: Our students built a model using a combination of train schedule information and turnstile data to detail the pattern of passenger movements. This included a model for both entry points and departure points for passengers in 15 minute blocks of time throughout the day. The result was a model that showed which times, and at which stations, there were passengers left on the platform after the arrival of a train. The transit authority logistics division was able to use this information to initiate a study of the train schedule
Problem: Like many organizations facing unprecedented amounts of data and reporting requirements, our leading Pharmaceuticals/Diagnostics client was working with many different legacy analytic tools, each with its own strengths. They realized they could be doing so much more with the data they had if they could organize their approach to data analysis in a more systematic way.
Solution: We tailored a private week-long training, based around R, that would help them meet their immediate and long-term needs. Having an instructor onsite allowed the team to ask questions based on actual experiences. We also designed class projects that directly and immediately benefited the group. By teaching them to use the R programming language as a central tool, they were able to collaborate with a common language, dramatically increasing efficiency and improving profits.