Skip to content

Knowledgebase for dynamic study, grading, and modules.

Use the references for the product loop, claim policy, and technical grounding behind AI-graded self-study.

How the learning engine works

Measurement and the mole

Significant figures, unit conversions, moles and formulas, and habits that catch calculation errors.

Reactions and stoichiometry

Balancing equations, limiting reactants, solution concentration, reaction types and redox.

Atoms, bonds and names

Electron configurations, periodic trends, Lewis structures, molecular shape and compound names.

Gases, energy and acids

Intermolecular forces, gas laws, enthalpy, Hess's law, and acids, bases and pH.

Atoms, light and bonding

Isotopes, atomic spectra, quantum numbers, electronegativity and orbital hybridization.

Solutions, phases and thermodynamics

Titrations, colligative properties, phase diagrams, entropy and free energy, and nuclear decay.

Kinetics and equilibrium

Rate laws, half-lives, activation energy, equilibrium constants and Le Chatelier's principle.

Equilibrium calculations and electrochemistry

ICE tables, weak acids and bases, buffers, solubility products and cell potentials.

Studying on your own

How practice adapts, how to plan it, and how to make it last.

Helping a learner

For teachers, tutors, parents, and mentors reading a learner's signals.

How we make claims

The sourcing behind any model or evaluation statement on this site.

How the learning engine works

  • Treats each answer as evidence, not a final score.
  • Pushes back when reasoning looks plausible but incomplete.
  • Uses the signal to choose the next question instead of restarting the track.

The product turns practice into a running model of understanding.

  1. A response becomes evidence.

    Answers, confidence, explanations, and visible reasoning suggest what might be solid and what might still be fragile.

  2. Evidence updates mastery signals.

    The system keeps a running, revisable estimate of concept-level understanding instead of treating one answer as a final score.

  3. The graph shapes the next move.

    Prerequisites and related concepts help decide whether to move forward, review a foundation, or ask a sharper follow-up.

Not a fixed course sequence.

Robotutor is adaptive, model-driven learning infrastructure: practice, mastery estimates, and follow-up questions adjust as your signals change.

  • Evidence-based mastery modeling

    Learner evidence is weighed across time, so guessing, uncertainty, and confident reasoning do not all look the same.

  • Graph-native sequencing

    Concepts are connected by prerequisite and related-idea edges, so a path can be derived from structure instead of a fixed lesson order.

  • Oppositional questioning

    Questions probe the reasoning behind an answer, including edge cases and changed assumptions, so misconceptions become visible.

  • Hybrid retrieval

    Semantic similarity and graph traversal are combined to ground responses in both meaning and explicit conceptual relationships.

For account help, launch feedback, or product questions, email support@precision-learning.org.