Workflow
Definition: A sequence of activities that software performs in a defined order.
Example: Collect data, check it, calculate an index and prepare a card.
Progressive path
Essential terms
From the word to the example
Each concept is presented through its meaning, an example and a useful boundary.
Automation
Definition: Automatic execution of defined activities or calculations.
Example: Update a report every morning without manually repeating the same steps.
Does not mean: It does not mean the system understands what it does.
RPA
Definition: Robotic Process Automation: a software robot repeats digital activities normally performed at a computer.
Example: Download a file, rename it, copy values and update a register.
Does not mean: It is not always machine learning.
Algorithm
Definition: A precise finite procedure for solving a problem or transforming data.
Example: Sort values, calculate an average or find a shortest path.
Does not mean: It is not automatically artificial intelligence.
Artificial intelligence
Definition: Methods enabling systems to classify, estimate, recognise or generate content.
Example: Compare several texts and prepare an initial summary for review.
Does not mean: It does not think or know the future.
Machine learning
Definition: Methods that learn relationships from examples or previous data.
Example: Use historical cases to detect anomalies or estimate consumption.
Does not mean: It does not guarantee the relationship will remain valid.
Model
Definition: A mathematical or logical relationship used to produce an estimate or classification.
Example: Estimate the probability of an event from previous data.
Does not mean: It does not know with certainty what will happen.
Forecast
Definition: An estimate about a future time or an event not yet observed.
Example: Estimate temperature, demand or probability over the next hours.
Does not mean: It is not a guarantee.
Classification
Definition: Assignment of a case to a category.
Example: Separate an invoice from a contract or normal data from anomalous data.
Does not mean: The category is not necessarily a decision.
Anomaly
Definition: A value or behaviour different from what the system considers usual.
Example: A sensor changes suddenly relative to its history.
Does not mean: It does not automatically mean an error or failure.
Data quality check
Definition: A check of presence, freshness, format, consistency and missing values.
Example: Block calculation when a source is stale.
Does not mean: It does not certify that the source is absolutely true or complete.
Human oversight
Definition: Human control and responsibility where interpretation or decision is required.
Example: Review a case classified as critical before acting.
Does not mean: It is not a purely formal step.
Next step
Terms you will meet at later levels
Baseline, feature, data drift, quality gate, idempotency and MLOps are explained in the glossary and technical pages when they become useful.
Ideas have a history
Algorithms, classification, inference and automation have antecedents in ancient mathematics, philosophy and mechanics.
Open the cultural path →