Lesson 5.4 Extraction

What best describes the structure of the World Wide Web?

• It is fully structured using scientific taxonomy.

• It is semi-structured using relational databases.

• It is largely unstructured, with some light organization like domain names.

• It is strictly organized by topic and keyword metadata.

Which of the following would be an example of structured data on the web?

• A random collection of blog posts.

• A webpage listing books in alphabetical order with author names and ISBNs.

• A series of image memes with no tags or captions.

• A forum with open-ended discussions.

Which tool is used by search engines like Google to index the web?

• Hackers

• Search filters

• Spiders (crawlers)

• Human editors

Which of the following is a challenge when processing large data sets?

• All data is pre-cleaned and ready to use.

• Data sets are always complete and unbiased.

• Cleaning, validating, and combining data from multiple sources.

• The data comes in uniform formats and structures.

What is the primary purpose of cleaning data before analysis?

• To delete unnecessary data permanently.

• To change the meaning of the data to suit our hypothesis.

• To make the data uniform without changing its meaning.

• To reduce the data size for storage purposes.

What issue might arise if you only survey students in Algebra I to represent the entire school?

• There’s not enough data.

• The data will be biased due to limited sampling.

• The responses will be too varied.

• It’s the best way to survey the school.

Which of the following statements about the Open Directory Project (DMoz) is TRUE?

• It was an automated web crawler developed by Google.

• It created a structured list of links organized by topic.

• It was designed only for government websites.

• It used spiders to rank web pages by popularity.

Why can big data be a problem even if it's valuable?

• It's illegal to store big data.

• It always contains biased results.

• Without adequate processing power, it can’t be effectively used.

• Most big data is already cleaned and processed.

What is one reason why Google search can produce “noisy” results?

• It doesn’t use algorithms to rank results.

• It uses structured directory systems exclusively.

• It includes personalized results based on user data.

• It only includes government-approved websites.

What is the best way to reduce bias in a data set?

• Increase the size of the biased data.

• Ignore any inconsistencies and process it anyway.

• Eliminate biased samples or combine with more diverse data.

• Only collect data from the most vocal users.