WGU D491 Exam Prep: Introduction to Analytics Practice

data analytics

The process of analyzing data to extract insights

descriptive analytics

A type of data analytics that summarizes past data

predictive analytics

The process of predicting future outcomes

prescriptive analytics

The process of recommending actions based on data

ETL

Extract, Transform, Load

big data

Data that is too large and complex to be processed by traditional data-processing software

data mining

To extract useful information from large datasets

unstructured data

A collection of social media posts

data warehouse

A system used for reporting and data analysis

regression analysis

A common technique used in predictive analytics

structured data

It is stored in a fixed format

data mart

To provide a subset of data for a specific business function

data visualization tool

An example is Tableau

data analytics process

The first step is data collection

data lake

A storage repository that holds a vast amount of raw data in its native format

descriptive analytics goal

To summarize past data

data cleaning technique

Data transformation

data analyst role

To analyze data and extract insights

business intelligence tool

An example is Power BI

data governance goal

To ensure data quality and compliance

cloud-based analytics benefit

Reduced data storage costs

data dashboard purpose

To provide a visual representation of key performance indicators

machine learning algorithm

An example is Linear regression

real-time analytics advantage

Faster decision-making

missing data handling method

Replacing missing values with the mean or median

hypothesis test purpose

To make inferences about a population based on sample data

categorical variable example

Gender

exploratory data analysis goal

To summarize the main characteristics of the data

statistical test for means

T-test

p-value indication

The probability of observing the data given that the null hypothesis is true

measure of central tendency

Mean

correlation coefficient purpose

To measure the strength and direction of the relationship between two variables

continuous variable example

Temperature

regression analysis purpose

To predict the value of a dependent variable based on one or more independent variables

Summarizing past data

To summarize past data

Predicting dependent variable

To predict the value of a dependent variable based on one or more independent variables

Cleaning and organizing data

To clean and organize data

Storing data securely

To store data securely

Healthcare application of data science

Predicting patient outcomes

Recommendation system in e-commerce

To predict customer preferences and suggest products

Finance application of data science

Predicting stock prices

Sentiment analysis in social media

To analyze and interpret the emotions expressed in text

Marketing application of data science

Predicting customer churn

Natural language processing (NLP)

To analyze and interpret human language

Transportation application of data science

Predicting traffic patterns

Anomaly detection in cybersecurity

To identify unusual patterns that may indicate a security breach

Retail application of data science

Predicting inventory needs

Machine learning in data science

To develop algorithms that can learn from and make predictions on data

Key principle of effective data visualization

Keeping the visualization simple and clear

Purpose of a scatter plot

To compare the relationship between two variables

Categorical data visualization

Bar chart

Advantage of using a dashboard

It allows for interactive exploration of data

Data storytelling

The practice of using data visualizations to convey a narrative

Purpose of a heat map

To display data density or intensity

Common tool for creating data visualizations

Tableau

Goal of data communication

To effectively convey insights and findings from data analysis

Example of a time series visualization

Line chart

Purpose of a pie chart

To display the proportion of categories within a whole

Type of supervised learning

Regression

Goal of unsupervised learning

To find hidden patterns or intrinsic structures in data

Example of a classification algorithm

Decision tree

Purpose of a confusion matrix

To evaluate the performance of a classification model

Common technique in dimensionality reduction

Principal component analysis (PCA)

Advantage of ensemble methods

They improve the accuracy and robustness of predictions

Example of a clustering algorithm

K-means

Purpose of cross-validation

To evaluate the performance of a model

Characteristic of a neural network

It is based on the structure and function of the human brain

Goal of reinforcement learning

To learn optimal actions through trial and error

Common evaluation metric for regression models

Mean Squared Error (MSE)

Purpose of feature scaling

To normalize the range of independent variables

Example of a supervised learning algorithm

Support vector machine (SVM)

Predict continuous values

To predict continuous values

Classify data

To classify data into distinct categories

Reduce dimensionality

To reduce the dimensionality of data

Visualize data

To visualize data

Prevent overfitting

Common method for preventing overfitting in machine learning models

Learning rate

To control how much the model's weights are adjusted with respect to the loss gradient

Unsupervised learning algorithm

Example of an unsupervised learning algorithm is K-means clustering

Random forest advantage

It reduces the risk of overfitting by averaging multiple decision trees

Activation function

Common activation function used in neural networks is Sigmoid

Validation set purpose

To tune the model's hyperparameters

Feature selection technique

Common technique for feature selection is Principal component analysis (PCA)

Convolutional neural network goal

To analyze and interpret images

Reinforcement learning algorithm

Example of a reinforcement learning algorithm is Q-learning

Dropout purpose

To prevent overfitting by randomly dropping units during training

Evaluation metric for classification

Common evaluation metric for classification models is Accuracy

Support vector machine advantage

It works well with high-dimensional data

Handling imbalanced datasets

Common technique for handling imbalanced datasets is Oversampling the minority class

Confusion matrix purpose

To evaluate the performance of a classification model

Deep learning framework

Example of a deep learning framework is TensorFlow

Generative adversarial network goal

To generate new data samples that are similar to the training data

Model selection technique

Common technique for model selection in machine learning is Cross-validation

Learning curve purpose

To evaluate the performance of a model over time

Boosting algorithm example

Example of a boosting algorithm is AdaBoost

Hyperparameter tuning goal

To optimize the performance of a model

Reducing overfitting technique

Common technique for reducing overfitting in neural networks is Using dropout

ROC curve purpose

To evaluate the performance of a classification model

Regularization technique example

Example of a regularization technique is Lasso regression

Convolutional neural network advantage

It is specifically designed to process grid-like data such as images

Evaluation metric for clustering

Common evaluation metric for clustering algorithms is Silhouette score