The experience of Quanticæ has been formalised into two methodologies that are used as the basis of our work. We have two different approaches: one for for Data Analytics & A.I. projects and another methodology adapted to consulting projects.


Data Analytics & Artificial intelligence

We propose a low risk approach based on piloting and incremental development. Our customers will invest once the suitability of the model for their business has been assessed

Ethics is our main asset

Quanticae has a strong commitment to ethics and to providing insights for the “good” not only for our customers but for the society as a whole. We execute our research with transparencyobjectivityindependence, and doing our best to minimize potential bias.

To this end, we strive to follow these principles:

  • The purpose and the scope of our deliverables will be clearly formulated so the objectives attempted to accomplish are understood.
  • Our studies will be contextually rooted and placed into a theoretical framework, so it is comprehensible how they relate and contribute to previous research and knowledge.
  • The approach and conceptual and methodological orientation will be appropriate for the problem and will be described. The inquiry logic will show how the methods and procedures selected are appropriate for accomplishing the study objectives.
  • When necessary various scenarios will be included in the analysis.
  • The collection and the selection of data will be adequately described, including how, when and for what purpose they were gathered.
  • Explicit user acknowledgment will be requested when collecting data from the users. The data will be anonymised as soon as it is possible, and the identifying data will be deleted.
  • Data will only be used for the purpose it was collected.
  • Limitations in the quantity or quality of data will be clearly stated.
  • Privacy is one of our top priorities. Only anonymous (or properly anonymised) data will be used in our analysis.
  • Our top priority is to avoid bias in our results, both in the causal and predictive analysis.
  • We will be very cautious when including any variable directly related to race, gender, religion, or other socio-demographic characteristic in our machine learning algorithms. We will only do it after an ethic assessment. Still, undetected proxy-variables can exist. When using machine learning, we provide auditing mechanisms and double-check the results to avoid any unintended bias.
  • The procedures and techniques used for analysis will be precisely and transparently described.
  • Evidences and facts will be presented in a way that can be verifiable and replicable by others.
  • It will be clear how outcomes of the analysis support claims and conclusions of the research.
  • In using machine learning we will do our best to explain the decision-making process behind the algorithms using the state-of-the-art technologies available.
  • Information about circumstances that may have implications for the validity or interpretation of results or conclusions will be included in the study.