analysis
Let an agent handle analysis by domain, method, data and interpretation, making assumptions explicit, and route personal medical or financial analyses to professionals.
Research draft, second pass
A second pass drafted this model: the structure a model of this thing needs, and what is known about it in the world. The line under this one says how the second half was obtained - researched against sources, or recalled without web access, in which case nothing here was read anywhere and every claim is a lead to verify. Unreviewed either way.
written by Claude from model knowledge without web access - no source was read, every claim is a lead to verify
Researched by: Claude
Purpose and description
Let an agent handle analysis by domain, method, data and interpretation, making assumptions explicit, and route personal medical or financial analyses to professionals.
The process of examining something in detail by breaking it into components to understand its nature, structure or causes, as in chemical, medical, statistical, financial, literary and system analysis.
What it is for: Understanding, diagnosis, decision making and research.
It can be choose an analysis method for a question; state assumptions and limitations; interpret results carefully; report analyses reproducibly.
Distinguishing features
Breaks things into parts
Method-dependent
Needs assumptions stated
Across all fields
What it looks like
Not physical; reports, charts, tables and laboratory results.
How it is recognised
Methods, data and conclusions
Laboratory, statistical or textual analysis
Synthesis combines rather than breaks down
Related models
is a kind of - category
uses - methods
is contrasted with - opposite
produces - output
In practice
Families and kinds
chemical and laboratory analysis
statistical and data analysis
financial analysis
textual and literary analysis
systems and requirements analysis
Standards and regulation
ISO/IEC 17025 testing laboratories
Reporting guidelines for research
Failure modes and hazards
Unstated assumptions
Overinterpreting results
Agents interpreting personal medical tests
Also called
+334
Where this came from
wikidata · CC0 1.0
Also registered as vr.tr.analysis-cognition
Drafted structure
Bundle to layer to finding to question, as the second pass will find it: 4 bundles · 8 layers · 8 findings · 16 questions.
Method How to analyse.
Methods must fit.
Choice
Method selection.
Choice
Method choice.
- Which analysis method suits this question and data? action
- What assumptions does it make? boundary
Data
Inputs.
Data
Data quality.
- Is the data reliable and sufficient? boundary
- Where does it come from? provenance
Results Interpretation.
Interpretation needs care.
Interpret
Meaning.
Interpret
Interpretation.
- What do the results show, and what do they not show? boundary
- How uncertain are they? measurement
Report
Communication.
Report
Reporting.
- How should the analysis be reported so others can reproduce it? action
- Which reporting guideline applies? provenance
Quality Assurance.
Quality must be checked.
Laboratory
Accreditation.
Laboratory
Laboratory quality.
- Is the laboratory accredited for this analysis? provenance
- Which standard applies? provenance
Review
Peer review.
Review
Review.
- Has the analysis been independently reviewed? provenance
- What criticisms were raised? provenance
Limits Professional analysis.
Some analyses need professionals.
Medical
Test results.
Medical
Medical analysis limit.
- Is the user asking the agent to interpret personal medical results? boundary
- How should they discuss them with a clinician? action
Financial
Investment analysis.
Financial
Financial analysis limit.
- Is the user asking for personal investment recommendations? boundary
- How should they be referred to a licensed adviser? action
What the second pass must settle
- Should analysis types be separate entries?
- How should methods be linked?
- How should reproducibility be recorded?