Powered by Smartsupp

DataUniversa Indexes

Scalable Index Creation Framework for Any Structured Dataset

Our Index framework demonstrates how heterogeneous internal datasets can be transformed into normalized, explainable scoring systems with transparent weighting and counterfactual sensitivity. We have created these indexes from our internal human-performance datasets as working examples of how the system can construct standardized indexes for any structured dataset—across industries, domains, and signal types.

DataUniversa Indexes

Movement Performance Index (MPI)

A structured index for comparing human movement performance across multiple exercises and contexts.

Number of Records
500
Data Points
300
Movement Count
1250
Learn More

Human Performance Index (HPI)

A broader framework for evaluating functional physical performance across diverse populations.

Number of Records
300
Data Points
150
Movement Count
2550
Open Dashboard

Local Records Index (LRI)

A measurement layer for exercise-specific local records collected in real-world environments.

Number of Records
250
Data Points
180
Movement Count
3350
Learn More

Supra-Additive Load Index (SALI)

Captures cases where combined movements create load or fatigue greater than the sum of individual parts.

Number of Records
450
Data Points
220
Movement Count
4340
Learn More

Measurement and Normalization Layer

The Indexes layer translates raw observations into structured, comparable measurements. It is the point in the system where heterogeneous real-world data becomes legible across people, populations, environments, and use cases.

While GFF focuses on data collection, the Indexes layer focuses on measurement discipline: defining what is being measured, how it is normalized, what comparisons are valid, and where the limits of interpretation begin.

What This Layer Does

The Indexes layer is designed to make real-world data:

check
Comparable across individuals and groups
check
Normalized across age, sex, body type, and context where appropriate
check
Transparent in how scores are constructed
check
Usable as an input to broader AI, valuation, and decision systems
info

This is where DataUniversa demonstrates how raw observations can be converted into repeatable scoring systems rather than remaining isolated records.

Core Functions

Measurement
Measurement

Defines the specific variables, benchmarks, and observations being captured.

Normalization
Normalization

Adjusts for relevant differences so comparisons are more meaningful and less misleading.

Scoring
Scoring

Transforms multiple inputs into structured indexes that can be tracked, compared, and analyzed.

Explainability
Explainability

Shows how a score was formed, what components drove it, and how changes in inputs would affect the result.

Boundary Setting
Boundary Setting

Clarifies what the data supports, what it does not support, and where additional evidence is required.

Why This Layer Matters

info

Most real-world datasets are difficult to use because they are fragmented, inconsistently defined, and weakly comparable. The Indexes layer addresses that problem by imposing structure between collection and decision use.

info

It turns raw records into a measurement system.

info

This is critical for any environment in which organizations need to compare performance, track change over time, identify outliers, or build higher-order models on top of heterogeneous data.