XTRACTIS®

the Reasoning AI for Trusted Decisions​ by design

Cognitive Artificial Intelligence

Transforms Data into Transparent Highly Predictive Intelligible Reliable Rational Explainable Auditable Certifiable Understandable Empowering Decision Systems.

Automatic Knowledge Extraction.

From data, XTRACTIS automatically discovers models of complex behavior in various domains. This can be human decisions, natural phenomena, or artificial processes, like control of autonomous devices. 

XTRACTIS-induced models come in the form of rule-based decision systems. They are ready-to-use offline or embedded, for prediction in real-time or for optimization.

Replicating human reasoning

To accomplish this challenge, our technology reproduces human induction to find relevant knowledge from observations as in experimental sciences. It also implements the other 2 modes of reasoning: Deduction and Abduction, to easily put this knowledge into use.

Reliable AND Transparent Model induction.

XTRACTIS models are robust: their performance is higher than mainstream AI techs, as benchmarks prove.

Models are also intelligible: humans are able to fully understand the internal decision-making logic of the Decision systems.

Experts can validate their models before pushing to end-use, or audit them before submission to certification to comply with AI regulations for high-risk applications.

Proprietary AI algorithms

XTRACTIS is not "just another AI" : its exclusive algorithms are based on the mathematical theory of Fuzzy Relations of order N proposed by Professor Zyed Zalila, founder and inventor. This theory is more consistent with real-world complexity than standard binary mathematics, while preserving human understanding.
Our solution also implements Collective and Evolving Learning algorithms to obtain models having the highest possible performance.

High-performance
Easy-to-use
AI Platform.

XTRACTIS comes in a no-code, feature-rich, all-in-one software platform accessible on a secure private cloud or on-premise servers.

for Models Intelligibility
# 5
for Models Robustness
# 5
Proprietary Algorithms
0 %
Code & Framework
100 %

Best Accuracy and Intelligibility
for multiple predictive applications.

As XTRACTIS is a Trusted General-Purpose AI, you can use it in your company or organization to:

Automate

complex decision processes to make them systematic, more objective, and more reliable.

Optimize

automated decisions to increase effectiveness, pro-activity, availability, productivity, and profitability.

Secure

decision-making support by using transparent models, to minimize risks when decisions are critical.

Embed

validated automated decision systems in autonomous devices, after their audit and certification.

You can always rely on the XTRACTIS-deduced decisions
thanks to the Robustness and Intelligibility of your models.

Whatever your sector, your application or your role.

XTRACTIS is game changing for
  • Researchers.
  • R&D Engineers.
  • Data Analysts.
  • Scientists.
  • Marketing Managers.
  • Domain Experts.
  • Decision Makers.

  • HEALTH / PHARMA

    Predictive Personalized Holistic Medicine: metabolic, epigenetic, physiological, pathological. Patient surveillance, Virtual Screening, Protein Homology, Toxicity, Drug Discovery, Health Insurance Risk Management...
  • BUSINESS / FINANCE

    Scoring and risk analysis, Econometrics, Malicious activities, Venture capital, Behavioral finance, Wealth management, Real estate finance, Strategy, Marketing & CRM, Customer behavior (appetence, attrition, emotions), HR, Legal...
  • INDUSTRY / R&D

    Product Design & Ergonomics, Sensory Marketing & Engineering, Smart Industry, Quality Control, Maintenance & Diagnosis, Logistics, Risk Analysis, Environment, Geomatics, Optimization, ADAS, Autonomous Vehicles...
  • DEFENSE / CYBER / SECURITY

    Autonomous Systems, Collaborative Devices, Command & Control, Malicious Activity Detection, Surveillance, Cybersecurity...

Many success stories
and use cases.

Personalized Medicine

XTRACTIS successfully diagnoses more than 98% of breast cancers by analyzing breast cell images, thus improving pathologists’ performance.

Product Design

XTRACTIS has been discovering the optimal formulations of blockbuster hygiene and wiping products from several major international brands since 2009. It helps adapt those products to the consumers’ sensory perception in each national market.

Military Surveillance

With a prediction error of only 1.6% in totally unknown situations, XTRACTIS wins the Tech-Challenge ‘Detection of objects’ organized by the French General Direction of Armaments and the French Military Intelligence Direction.

Scoring & Risk Analysis

XTRACTIS succeeds in providing predictive models for credit default, improving the Gini coefficient by 10+ points compared to the classical model: less capital to immobilize and better detection of the riskiest files, with an intelligible model as imposed by european regulators.

Our public Use Cases include comprehensive benchmarks against Machine Learning techniques, such as Neural Networks, Boosted Trees, Random Forests, and Logistic Regression.

These studies illustrate the ability of XTRACTIS to automatically induce knowledge from real-world data in the form of predictive and intelligible mathematical relationships.

For each application, we also show how this knowledge is used to obtain explained predictions for situations unknown from the learning data set.

No-Code All-in-One AI Platform.

The XTRACTIS AI is working its magic inside an all-in-one software suite, with 4 powerful interdependent softwares. We call them our robots! Meet REVEAL, PREDICT, OPTIMIZE and MONITOR.

Induction Robots

Discover your process rules from data and understand how it behaves.

XTRACTIS REVEAL

Deduction Robots

Make real-time predictions at high-frequency for new input situations.

XTRACTIS PREDICT

Abduction Robots

Automatically invert your models to discover the most optimal solutions in no time.

XTRACTIS OPTIMIZE

Supervision Engine

Supervise the quality of your deployed models to take action when they expire.

XTRACTIS MONITOR

Prof. Zyed ZALILA

CEO-founder, Intellitech

Cognitive psychology studies show that conscious human reasoning is limited to at most 9 criteria simultaneously. As a result, the non-augmented human will not be able to design decision-making rules, involving tens or even hundreds of interacting variables, to model complex behaviors.
Thanks to its Augmented Fuzzy AI, XTRACTIS is positioned as a universal predictive modeler of complex processes.
Like an exobrain that extends the human cognitive capacities, it allows us to apprehend all the pluridimensionality of the real world, without reduction nor simplification.
A human being does not reason in black or white, but with an infinite palette of shades of gray. Human reasoning cannot be modeled by a binary logic, but thanks to infinite-valent quantum logic, which we commonly call "fuzzy logic".
In a real world that we perceive with vagueness, uncertainty and subjectivity, using Fuzzy Mathematics allows modeling of complex phenomena and processes in a nuanced and more accurate way.
Paradoxically, XTRACTIS can explain a human expert's unconscious implicit knowledge which, in essence, cannot be explained. XTRACTIS can also decipher the hidden decision strategy of a black-box model.
Previous slide
Next slide

They have trusted XTRACTIS

From French public institutions to renowned international companies, many have trusted XTRACTIS. The future of AI is now!