Google Advanced Data Analytics cheat sheet
Google Advanced Data Analytics · Keep the key ideas within reach.
Data projects, communication and accountable business value
Begin from an actual decision and measurable outcome.
Preserve data lineage, definitions, scope and limits.
Communicate uncertainty and practical tradeoffs alongside statistics.
A pilot, course certificate and production system establish different things.
Python and dataframe reasoning for applied analytics
Python and dataframe types affect transformation meaning.
Labels, positions, alignment and join multiplicity need explicit checking.
Notebook state does not replace reproducible execution.
Verify small-table results, missing-value choices, units and populations.
Statistical inference, regression and model assumptions
Sampling/design assumptions constrain inference.
Confidence intervals and p-values have different precise interpretations.
Regression association is not automatically causal.
Prevent leakage and select evaluation measures for the actual decision.
Intelligent search, learning and evaluation
Define states, actions, objective and goal test before choosing a search.
Breadth-first minimum-edge guarantees need equal edge costs.
A-star guarantees depend on heuristic and repeated-state handling assumptions.
Separate training, validation and final evaluation roles.
Precision and recall use different denominators; accuracy can conceal class imbalance.
Data, Databases, and Information Retrieval
Primary keys are unique and non-null; foreign keys enforce referential integrity.
WHERE filters rows before grouping; HAVING filters aggregated groups.
COUNT(*) counts rows; COUNT(column) skips NULL values.
LEFT JOIN preserves unmatched left rows; INNER JOIN removes them.
Precision = relevant retrieved / retrieved; recall = relevant retrieved / all relevant.
Identity, access and cryptographic protection
Identification, authentication, authorisation and accounting are four distinct steps.
MFA needs two different factor categories, not two instances of one.
RBAC suits stable job roles; ABAC suits context-dependent, fine-grained rules.
Kerberos uses a KDC with TGT and service tickets for SSO.
Symmetric ciphers handle bulk data; asymmetric ciphers solve key distribution and signatures.
Signing uses a specified signature scheme with the signer’s private key; verification and custody are essential to origin evidence.
ECB leaks patterns; use authenticated modes such as GCM or encrypt-then-MAC.