[NEW] SAS Certified Professional: AI & Machine Learning

free certified natural language processing nlp course & Computer Vision (30%): Deep dives into text preprocessing, sentiment analysis, and the complexities of image feature extraction and object classification.

Forecasting & Optimization (30%): Understanding time series data, developing accurate forecast models, and formulating complex optimization problems for business solutions.

B. Target encoding or binning based on the response variable.

C. Removing the variable entirely to reduce noise.

D. Using a simple label encoder (1, 2, 3...).

E. Duplicating the rows to balance the classes.

F. Replacing all values with the mean of the dataset.

Correct Answer: B

Explanation:

A (Incorrect): One-hot encoding creates a column for every level, leading to the "curse of dimensionality" when cardinality is high.

C (Incorrect): High-cardinality variables often contain significant predictive power; removing them should be a last resort.

D (Incorrect): Simple label encoding implies an ordinal relationship that may not exist, potentially misleading the model.

E (Incorrect): This is a technique for class imbalance, not for handling categorical cardinality.

F (Incorrect): You cannot calculate a mean for categorical text data.

Question 2: In a Natural Language Processing (NLP) workflow using SAS, what is the primary purpose of "Stemming" during the text preprocessing phase?

B. To translate the text into a different language.

C. To reduce words to their root form by removing suffixes.

D. To count the total number of unique words in a document.

E. To fix spelling errors automatically.

F. To encrypt the data for secure transmission.

Correct Answer: C

Explanation:

A (Incorrect): This describes Sentiment Analysis, which happens after preprocessing.

B (Incorrect): Stemming is a linguistic reduction tool, not a translation tool.

D (Incorrect): This describes token frequency or vocabulary counting.

E (Incorrect): Spell-checking is a separate preprocessing step; stemming assumes the word is spelled correctly.

F (Incorrect): Stemming is for analysis, not security or encryption.

Question 3: A data scientist is building a forecasting model in SAS Viya. If the time series data exhibits a clear seasonal pattern that increases in magnitude as the trend increases, which model type is most appropriate?

B. Additive Seasonal Model.

C. Multiplicative Seasonal Model.

D. Moving Average with a window of 2.

E. Random Walk without drift.

F. K-Means Clustering.

Correct Answer: C

Explanation:

B (Incorrect): Additive models assume the seasonal component is constant regardless of the trend.

A (Incorrect): Linear regression does not inherently capture seasonality without significant feature engineering.

D (Incorrect): A simple moving average is a smoothing technique, not a forecasting model for complex seasonality.

E (Incorrect): A random walk assumes no predictable pattern, contradicting the premise of a "clear seasonal pattern."

F (Incorrect): Clustering is an unsupervised technique for grouping, not for time series forecasting.

You can retake the exams as many times as you want.

This is a huge original question bank.

You get support from instructors if you have questions.

Each question has a detailed explanation.

Mobile-compatible with the Udemy app.

30-days money-back guarantee if you're not satisfied.