Which ML Algorithm Classifies Genes into 20 Categories with Transparent Inner Mechanisms?
A company wants to classify human genes into 20 categories based on gene characteristics. The company needs an ML algorithm to document how the inner mechanism of the model affects the output. Which ML algorithm meets these requirements?
Community Votes
100% of anonymous learners picked answer A. Votes are pick records left by other test-takers — they are not the verified answer.
Community Insight
This question tests the dual requirement of multi-class classification capability and model interpretability, where decision trees uniquely satisfy both by exposing their internal split logic.
Decision trees are the ideal ML algorithm for multi-class classification tasks like categorizing human genes into 20 groups while providing transparent, interpretable decision paths. They allow practitioners to document exactly how input features influence outputs, unlike black-box models.
Candidates often choose logistic regression, mistakenly believing it handles multi-class problems well, but standard logistic regression is inherently binary and lacks the transparent tree-based decision path that decision trees offer.
Community Discussion (5 comments)
Comments & Corrections
No comments yet — spotted an error or have a note? Share it below.
Expert Analysis
Why Decision Trees Are the Correct Answer
The question presents two critical requirements:
1. Multi-class classification — categorizing genes into 20 distinct categories. 2. Interpretability / Transparency — the ability to document how the inner mechanism of the model affects the output.
Decision trees excel on both fronts. They natively support multi-class classification by splitting nodes based on feature thresholds (e.g., gene expression levels) until a leaf node assigns one of the 20 categories. More importantly, decision trees are white-box models: you can trace every prediction from root to leaf, documenting exactly which gene characteristics drove the classification at each step. This makes them ideal for regulated or scientific domains where model explainability is mandatory.
Why the Other Options Fail
- B. Linear regression — This is a regression algorithm, not a classification algorithm. It predicts continuous numerical values, not discrete categories.
- C. Logistic regression — Standard logistic regression is designed for binary classification (two classes). While multinomial logistic regression exists, it does not provide the same transparent, step-by-step decision documentation that a decision tree's split path offers. Community members noted this trap: "Logistic regression is for binary classification and not suitable for 20 categories."
- D. Neural networks — While neural networks can handle multi-class classification, they are black-box models. Their inner mechanisms (weights, activations across hidden layers) are extremely difficult to interpret and document in a human-readable way, directly violating the transparency requirement.
Key Takeaway
When an exam question emphasizes both classification into multiple categories and explainability of the model's inner workings, always lean toward decision trees or other inherently interpretable models like rule-based systems.
Official Reference
Exam Strategy
When you see keywords like 'document inner mechanism,' 'transparent,' or 'explainable,' immediately eliminate black-box models (neural networks, ensemble methods). Then verify whether the remaining interpretable models support the required task type (classification vs. regression, binary vs. multi-class).
Related Analysis
Practice All AIF-C01 Questions
Access 100 questions with complete answers and detailed explanations.
View Full AIF-C01 Practice Test →