PredReX- Predict and recommend like a king with Excel©

What PredReX is about:

- Predict time series using long short-term memory (LSTM), a special kind of recurrent neural network (RNN) used in the field of Deep Learning (DL).

- Recommend items based on data mining techniques like pattern recognition, correlation mining and association rule learning.

PredReX unleashes state of the art analytics in a simple to use Excel© (owned by Microsoft Corporation) template. Beside Excel©, no further installations are necessary. You can easily adapt algorithm's parameters and directly benchmark the results. We moved all the Machine Learning (ML) coding into the background so you can focus on what matters most: extracting key insights out of your data!

PredReX: PREDict and Recommend with EXcel like REX (latin for king)!

How PredReX works

Input your eg. sales data into the PredReX Template:

Beside sales data you can also use all other kind of data which you want to predict on a time series or want to conduct pattern recognition on. Eg. take your error log file data to predict the next upcoming failure.

Run any of these algorithms: instantly analyse your results in an easygoing tabular format.

That's how results for Market Basket could look like:

Spot time series prediction including loss function depending on your chosen error metric:

Or receive individually chosen top sales item recommendations per customer based on item-customer similarity:

Furthermore you can conveniently tinker with lots of parameters:

... to finetune your individual data needs:

How to get PredReX

To start your time series and recommender analytics in Excel© you can download the PredReX template right here. For just 19.99 USD you receive one support hour for free for possible data preparation or result interpretation.

Any modification requests, please let us know. We work hard to keep you 100% satisfied.

If you're interested in the theory behind these algorithms you might want to check out our book Data Driven Dealings Development on Amazon.