NXG Logic offers an industry-leading portfolio of analytics and teaching software that empowers companies and universities of all sizes to conquer their most complex research challenges. From the laboratory bench to the patient's bedside, improving efficiency and productivity is in our DNA and our software. This unmatched experience is why NXG Logic is a leader in data analytics. Designed for the Cambrian explosion in ML/AI, the Explorer software platform delivers a hybrid of mixed analytics to rapidly thrust forward the discovery of novel results, which are automatically formatted for presentation and publication. At NXG Logic, we're enabling researchers to expand their horizons and improve their productivity. For more information, visit NXG Logic.
Our programming efforts focused on development of several fast-formatting technologies which combine output from runs made on multiple variables, as a way to overcome difficulties associated with the majority of current-day approaches. The realization was that employment of most of the popular software packages (SAS, SPSS, Stata, Matlab, Statistica, R-Project, etc.) for combining results of analyses on many variables was much too time-consuming, since each variable needed to be tested independently, and the the results had to be transposed into a presentable/publishable format. This meant that it could take hours to perform 2- and k-sample tests and model building strategies for regression (and diagnostics) for 15-20 continuous and categorical variables.
Our innovation has led to technologies which result in a significant increase in productivity due to the reduced time required for generating presentation/publication-ready formatted outputs, which rapidly thrusts forward a user's ability to interpret and understand results faster and more efficiently.
The first order of business was to develop parallel feature analysis (PFA), so that a user could simultaneously generate summary statistics, Shapiro-Wilk tests, histograms, and count frequencies for multiple continuous and categorical features.
A list of our fast-formatting technologies implemented in the Explorer package are listed below:
PFA - Parallel Feature Analysis
FFOSS - Fast Formatted Output for Summary Statistics
FFOMT - Fast Formatted Output for Multiple Tests
FFORM - Fast Formatted Output for Regression Models
FFOA - Fast Formatted Output for Association
FFOCD - Fast Formatted Output for Class Discovery
FFOCP - Fast Formatted Output for Class Prediction
Portability Not Production Deployment
Hypothesis-Generated Data and Not Fishing Expeditions
NXG Logic software was not developed for deployment on HPC clusters (servers) in order to routinely process large-scale data in a pipeline. Instead, it was developed for portability, and independence from servers, command line code, and batch jobs.
NXG Logic Explorer was also designed for use with smaller experimentally-derived hypothesis-generated datasets consisting of, for example, several hundred to a thousand objects (records, cases, samples, etc.). As such, Explorer is not intended for fishing expeditions to find signals or patterns in large-scale administrative or production datasets which were not generated specifically as a result of hypothesis-driven research. Data generated from hypothesis-driven research undergoes human-based case review by committee and quality checks for each object before the data are entered into a database. (QC checks commonly include range checking, if-then logical, character vs. digit filtering, etc.) In addition, the return on investment (ROI) from "fishing with dynamite" within marginal probabilities is usually low, since patterns that do emerge often disappear as soon as conditional probabilities are considered.
NXG Logic's mission is to provide accurate, user-friendly, fully-supported software packages for research and operations. The company is led by Dr. Leif Peterson.