1️⃣ Question 1

Purpose of GANs?

• ❌ Coherent text

• ❌ Music generation

• ❌ High-quality images (GANs do this, but not the primary definition)

• ✅ Creating realistic data samples

Explanation:
GANs generate realistic synthetic data, including images, videos, and more.

---

2️⃣ Question 2

How does generative AI help data science professionals?

• ❌ Synthesize medical images

• ❌ Generate game levels

• ❌ Generate text

• ✅ Augment datasets with synthetic data

---

3️⃣ Question 3

Tool for image data augmentation:

• ❌ Dialogflow

• ❌ Autoencoders

• ❌ Magenta

• ✅ CycleGAN

---

4️⃣ Question 4

How generative AI helps in idea generation during problem definition:

• ❌ Filling missing values

• ❌ Monitor drift

• ❌ Generate adversarial cases

• ✅ Mimicking existing product descriptions / marketing content

---

5️⃣ Question 5

Generative AI model good for sequential data:

• ❌ Flow-based

• ❌ VAEs

• ✅ Autoregressive models

• ❌ GANs

---

6️⃣ Question 6

Model generating coherent text (poetry, scripts, emails):

• ❌ VAEs

• ✅ Autoregressive models (e.g., GPT)

• ❌ Flow-based

• ❌ GANs

---

7️⃣ Question 7

Model that generates data adhering to original distribution:

• ❌ GANs

• ❌ VAEs

• ✅ Flow-based models

• ❌ Autoregressive

---

8️⃣ Question 8

Data consideration for generative AI:

• ❌ Interpretability

• ❌ Bias mitigation

• ❌ Ethical mechanisms

• ✅ Privacy, encryption, access control

---

9️⃣ Question 9

Tool for semi-structured data augmentation (text descriptions & code):

• ❌ StyleGAN2

• ❌ CTGAN

• ❌ SDV

• ✅ Copilot

---

🔟 Question 10

Correct prompt for the SQL update:

• ✅ Replace the zero values in the ZN column with NULL.

• ❌ Find null values

• ❌ Find zero rows

• ❌ Update all values

---

1️⃣1️⃣ Question 11

How generative AI improves imputation:

• ❌ Outlier detection

• ❌ Language translation

• ❌ Latent code

• ✅ Learns patterns and generates plausible missing values

---

1️⃣2️⃣ Question 12

Cultural challenge:

• ❌ Data quality

• ❌ Lack of standardization

• ❌ Copyright

• ✅ Trust and transparency

---

1️⃣3️⃣ Question 13

Visualization to verify outliers:

• ❌ Color coding

• ❌ Histograms

• ❌ Annotation

• ✅ Box plots

---

1️⃣4️⃣ Question 14

Open-source AutoML library:

• ✅ AutoGluon

• ❌ H2O Driverless AI

• ❌ Vertex AI

• ❌ DataRobot

---

1️⃣5️⃣ Question 15

Analysis to identify patterns for deeper investigation:

• ❌ Feature engineering

• ❌ Bivariate

• ❌ Univariate

• ✅ Hypothesis generation

---

1️⃣6️⃣ Question 16

Model consideration technique improving interpretability:

• ❌ Imaging data

• ✅ Feature attribution

• ❌ Perpetuate biases

• ❌ Manipulative inputs

---

1️⃣7️⃣ Question 17

Organizational challenge:

• ❌ Continuous learning

• ❌ Data quality

• ❌ Risk aversion

• ✅ Change management

---

1️⃣8️⃣ Question 18

Simulation & data-augmentation tool:

• ❌ Jukebox

• ❌ Autoencoders

• ✅ Unity ML-Agents

• ❌ StyleGAN

---

1️⃣9️⃣ Question 19

Anomaly-detection generative AI tool:

• ❌ DALL•E

• ❌ CycleGAN

• ❌ Dialogflow

• ✅ Isolation Forest

---

2️⃣0️⃣ Question 20

How generative AI helps in data exploration & preparation:

• ❌ Generate new features

• ❌ Recommendations

• ❌ Detect anomalies only

• ✅ Data augmentation: fill missing values, handle outliers, generate synthetic data

---

🧾 Summary Table

---