Data Science Analysis in Business Operations Quiz

ExcitedPelican avatar

Start Quiz

Study Flashcards

30 Questions

What is the main purpose of using customer comments about a product in customer service departments?

To identify aspects for improvement

How are data science tools applied in smart cities?

To transform data into actions for residents' benefit

What is the first dimension of data science activities according to the text?

Data Flow

What is the main goal of data curation in the context of data science activities?

To refine collected data

In which field have recent advances in artificial intelligence allowed diagnosis of diseases when specialists are not available?

Medical applications

What does the storage structure of data aim to achieve in the context of data flow?

Transparency, completeness, and accessibility

What is one of the important aspects of data science mentioned in the text?

Extracting actionable insights

How did the United Parcel Service (UPS) reduce fuel usage and miles off its routes?

By installing sensors in vans and combining data with GPS information

What does the Internet Movie Database (IMDB) provide online?

Data about all elements of the movie industry

In a supermarket scenario, what do managers gain solid knowledge of by applying data science elements?

Costs and revenues

What type of information does IMDB aim to extract using data science tools?

Information about actors in highest-rated movies

How did the application of data science tools benefit UPS according to the text?

Reduced fuel usage and miles off routes

What does the area under the ROC curve (AUC) measure?

Efficiency of the model

Why is it important to translate the output of a regression prediction model into a number?

To make it easier to understand

What does the absolute error measure in evaluating a regression model?

Difference in model's output and desired output

How is relative error calculated in relation to absolute error?

(d-y)/d * 100 %

Why is it mentioned that relative error may not be quite representative for small numbers?

Small numbers may lead to invalid operations

Which of the following is NOT a typical metric used to evaluate a regression model?

Classification error

What is the purpose of developing a model with a feedback loop that can accommodate changes like product price adjustments?

To increase the accuracy of the model's predictions.

In the context of building an intelligent model, what role does the model's confidence in its predictions play?

It influences end users' actions without verification.

Why is the 'machine learning canvas' tool helpful in identifying use cases?

To provide a user-friendly procedure for business managers.

What does the 'machine learning canvas' tool aim to achieve for business managers?

Consolidating all steps needed to identify use cases and their value propositions.

In what scenario could a prediction model be automatically accepted or rejected without contacting an end user?

When the model's confidence in its predictions is high.

How does including a feedback loop in a model help in accommodating changes like product price adjustments?

By allowing for model retraining based on new data.

What is the purpose of feature selection in machine learning?

To select informative and relevant features by applying correlation analysis

Why is it important for features to have a low degree of intercorrelation with other features?

To make the data more understandable and avoid redundancy

What role does a domain expert play in feature selection?

They guide the process and review the list of suggested relevant features

What is the main purpose of developing a learning mathematical algorithm in machine learning?

To extract knowledge from data and predict future outcomes

Which type of analytics is used to understand underlying data patterns in machine learning?

Descriptive analytics

How is the learning technique determined in machine learning?

By choosing between unsupervised and supervised learning based on the nature of the problem

Study Notes

ROC Curve and Evaluation Metrics

  • A ROC curve measures the efficiency of a model, with the ideal model being closest to the upper left corner.
  • The area under the curve (AUC) is a measure of efficiency, with an ideal model having an AUC of 1.
  • Regression model evaluation metrics include:
    • Absolute error: the absolute difference between the model's output and the desired output.
    • Relative error: the absolute error normalized with respect to the desired output to obtain a unit-less percentage.

Data Science Applications

  • Data science is used to extract useful information and make predictions in various industries, such as:
    • Supermarkets: to analyze costs and revenues and predict outcomes of different business scenarios.
    • Logistics: UPS used data science to reduce fuel usage by 8.4 million gallons and shave 85 million miles off its routes.
    • Entertainment: IMDB uses data science to extract information and answer questions about the movie industry.
  • Data science tools can be used to:
    • Identify customer satisfaction and areas for improvement in customer service departments.
    • Improve public transportation systems in smart cities.
    • Diagnose diseases in medical applications using artificial intelligence.

Data Science Activities

  • Data science activities are conducted in three dimensions: data flow, data curation, and data analytics.
  • Data flow involves collecting, storing, and managing data.
  • Data curation involves refining collected data, including handling changes to data.
  • Data analytics involves extracting insights and making predictions from data.

Machine Learning

  • Machine learning is used to extract knowledge from data and make predictions.
  • Types of machine learning include:
    • Unsupervised learning: using cluster analysis to learn from data.
    • Supervised learning: using classification and regression approaches to learn from data.
  • Machine learning can be used to:
    • Automate decision-making processes.
    • Identify use cases and achieve value propositions using tools like the machine learning canvas.

Feature Selection

  • Feature selection involves selecting informative and relevant features from a dataset.
  • Correlation analysis is used to separate redundant features and keep features that show high correlation with the target variable.
  • The result is a reduction in feature sets and more comprehendable data.

Test your knowledge on how data science analysis helps extract useful information and insights for business operations. Learn how data is presented to managers for decision-making in areas like costs, revenues, and future expectations.

Make Your Own Quizzes and Flashcards

Convert your notes into interactive study material.

Get started for free

More Quizzes Like This

Use Quizgecko on...