Analytics Explorer 2023.1

Overview

Using S&P Global's vast data stores and partnering with TIBCO Spotifire for the visualizations, Analytics Explorer provides advanced data analysis, data preparation, and powerful prediction capabilities. Machine learning models allow for robust multivariate analysis for application to lithologies, production, completions, field development, and others. S&P Global's predictive models provide insight into the how and why behind reservoir and well production.

Analytics Explorer can currently connect to the Kingdom and Harmony Enterprise databases, Enterprise Data Management (EDM) databases, standalone SQL and PostgreSQL databases. All databases (with the exception of SQL Server Express, a Kingdom-specific lightweight SQL Server database) require you to enter credentials, unless you are opening Analytics Explorer from inside the application (and you have already entered credentials to open the project).

Open Spotfire directly or connect to Spotfire through the application, currently Kingdom and Harmony Enterprise. In Spotfire, the Analytics Explorer menu includes the following:

Harmony

Connect to a Harmony project database. See Harmony Analytics Explorer

Kingdom

Connect to a Kingdom project database. See Kingdom Analytics Explorer

S&P Global Content

EDM/SQL/PostgreSQL - connect to an existing database
EDIN data - access exploration, production, and midstream information and analysis spanning over 250 countries.
EDIN Reservoir Benchmarking - provides the tools and data to compare your assets against those of similar complexity using a machine learning algorithm developed to create a model on global reservoir data.
Impact data - Impact North American data comes from multiple sources, including state registries, and S&P Global data stores. The data is cleaned and curated for accuracy and completeness.
Impact Predict - download S&P Global Impact basin data for advanced analytics.

Data Preparation

Auto Imputation—impute missing values using machine learning and selected input data.
Principal Component Analysis (PCA)— identify patterns in data, and express the data in such way as to highlight their similarities and differences.
Multicollinearity Analysis— run as an independent data preparation step or as part of the Machine Learning Predictions workflow. Multicollinearity Analysis helps determine the best inputs for the machine learning model, or conversely, the attributes to exclude from the model for optimal results.

Machine Learning

Cluster Analysis —Cluster Analysis or Clustering is the process of dividing the population or data points into a number of groups with data points in the same group having similar user-selected attributes or characteristics.
Machine Learning Predictions (MLP) - Machine Learning is based on the idea that systems can learn from data, identify patterns, and make decisions with minimal intervention. Given a set of input data, the algorithm makes predictions for specified field(s).
Run Machine Learning Model - run the machine learning model you created in Machine Learning Predictions on other tables.

Table-Specific Calculations

In addition to the menu items above, a number of calculations are available from the right mouse button menu when in certain tables. See Table-Specific Calculations

Please see the latest Analytics Explorer release notes for compatibility metrics.

TIBCO issues periodic security advisories with their releases. For additional information, see TIBCO Security Advisories.