MS365 Business Analytics & Business Intelligence

Podcast

Play an AI-generated podcast conversation about this lesson

Questions and Answers

[Blank] analytics involves using quantitative methodology to make data-driven decisions, ranging from descriptive statistics to complex data mining.

Business

[Blank] intelligence primarily focuses on data aggregation, visualization, and reporting to understand who, what, when, where, and why.

Business

Business analytics includes methodology that moves past explanatory modeling and can be used to create ______ models used for business decision making.

predictive Signup and view all the answers

[Blank] mining is where statistics and machine learning meet, which can be used to cluster customers into personas.

Data Signup and view all the answers

The four V's of Big Data are Volume, Veracity, ______, and Velocity.

Variety Signup and view all the answers

[Blank] science is a mix of skills in statistics, IT, math, programming, and business.

Data Signup and view all the answers

[Blank] Intelligence (AI) is the science of engineering and making intelligent machines.

Artificial Signup and view all the answers

[Blank] and Feinzig are the authors of the text.

Guenole Signup and view all the answers

[Blank] AI is a type of deep learning used to create new content.

Generative Signup and view all the answers

[Blank] are a type of Machine Learning Model that is typically used for prediction.

Regression Models Signup and view all the answers

The Dartmouth Summer Research Project on Artificial Intelligence convened to solve the problem of [Blank].

AI Signup and view all the answers

John McCarthy, a professor at Dartmouth, Stanford, MIT, and Princeton, created the programming language [Blank].

LISP Signup and view all the answers

Alan Turing is best known for his seminal paper `Computing Machinery and Intelligence`, published in [Blank].

1950 Signup and view all the answers

Alan Turing created the Bombe Machine, a code breaking machine, during [Blank].

WWII Signup and view all the answers

The [Blank] Test measures a machine's intelligence by evaluating its ability to exhibit human-like conversational behavior.

Turing Signup and view all the answers

The Turing test involves an interrogator trying to distinguish between a real agent and a ______ pretending to be the real agent.

machine Signup and view all the answers

______ AI, also known as weak AI, focuses on creating models that perform one specific task.

Narrow Signup and view all the answers

While the math behind many modern deep learning algorithms is decades old, their practical application was initially limited by the lack of sufficient computer ______.

hardware Signup and view all the answers

______'s Law states that the number of transistors in an integrated circuit doubles approximately every two years.

Moore Signup and view all the answers

The increasing density of hard drives is described by ______'s Law, which states that storage density increases at a rate faster than Moore's Law.

Kryder Signup and view all the answers

The challenge of handling intellectual property issues is particularly relevant in the field of ______ AI, where models can generate novel content.

Generative Signup and view all the answers

Distinguishing between predictive modeling and ______ modeling involves understanding whether the goal is to forecast continuous values or assign data points to predefined categories.

classification Signup and view all the answers

______ AI is a theoretical design that would allow AI models to be created to perform a broad array of general actions

General Signup and view all the answers

Which of the following is NOT considered a core idea of data mining?

Optimization (A) Signup and view all the answers

A multi-class classification model can only assign an observation to one class.

True (A) Signup and view all the answers

What is the primary difference between classification models and predictive models, in terms of output?

Classification models produce categorical outputs, while predictive models produce continuous numerical outputs. Signup and view all the answers

In a _____________ classification model, an observation can belong to multiple classes simultaneously.

multi-label Signup and view all the answers

Which type of classification model would be most suitable for determining if an email is either 'Important', 'Normal', or 'Low Priority'?

Multi-Class Classification (A) Signup and view all the answers

Predictive models produce categorical outputs similar to classification models.

False (B) Signup and view all the answers

Match the classification model type with its appropriate use case:

Binary Classification = Determining if a transaction is fraudulent or not Multi-Class Classification = Identifying the species of a plant from a set of known species Multi-Label Classification = Identifying all objects present in an image (e.g., car, pedestrian, bicycle) Signup and view all the answers

A self-driving car needs to identify multiple objects in its surroundings simultaneously, such as pedestrians, traffic lights, and other vehicles. Which type of classification model is most suitable for this task?

Multi-Label Classification (C) Signup and view all the answers

Which of the following actions is least likely associated with the 'Amount of' considerations in a business context?

Customer satisfaction rating (C) Signup and view all the answers

Data exploration and visualization are primarily used for creating predictive models rather than understanding and cleaning data.

False (B) Signup and view all the answers

Briefly explain the primary goal of dimension reduction in data mining.

improving model performance by reducing variables with weak predictive power Signup and view all the answers

In the SEMMA data mining workflow, the 'M' stands for ______.

modify Signup and view all the answers

Which of the following is a key difference between Supervised and Unsupervised learning techniques?

Supervised learning uses labeled data for training, while unsupervised learning does not. (A) Signup and view all the answers

Match the following data mining workflow stages with their primary objective:

Business Understanding = Define project objectives and requirements Data Preparation = Clean and transform data for modeling Modeling = Apply selected techniques to build predictive models Evaluation = Assess model performance against project goals Signup and view all the answers

Which of the following is the best description of the CRISP-DM methodology's 'Data Understanding' stage?

Analyzing data to identify patterns and insights (D) Signup and view all the answers

Data reduction always leads to improved model performance by simplifying the dataset and removing noise.

False (B) Signup and view all the answers

Flashcards

Classification Models

AI models that predict categories or classes.

Regression Models

AI models that predict continuous numerical values.