AP CSP Data (1)

Computers are described as processing data iteratively. In what order will the computer process the following program?

Line 1: a ← 38

Line 2: b ← 54

Line 3: a ← b + a

Line 4: DISPLAY(a + b)

• A computer will process all lines of code at once.

• A computer will process the metadata before the pure data.

• A computer will process data one step at a time in the given 1, 2, 3, 4 order.

• A computer will display the value 92 while processing lines 1, 2, 3, and 4 at the same time.

Answer explanation

 “Iterative” refers to a repeated process, which is how computer code is processed. Each piece of data is put through a process, which is repeated for every piece.

(A) would require multiple lines of code to be evaluated at once. To calculate line 3 correctly, lines 1 and 2 would have needed to be evaluated first.

(B) is dependent on the program in question (most don’t even require metadata).

(D) is the value of a + b without changing the value of a.

An Alaskan biologist is tracking a pod (group) of whales using tracking collars. For the whales, the following geolocation data are collected at frequent intervals:

■Time

■Date

■Geographical location of the sea mammals

Which of the following questions about a whale could NOT be answered using only the data collected from the tracking collars?

• Approximately how many miles did the animal travel in one week?

• Does the animal travel in groups with other tracked animals?

• Do the movement patterns of the animal vary according to the weather?

• In what geographic location does the animal typically travel?

Answer explanation

Searching for patterns in big data can be used in careers other than strictly computer science such as Marine Biology. Tracking whales is one such example. Because the data are tracking multiple whales, it would be possible to know if a particular whale was traveling with other whales, the typical travel patterns, and how many miles the animal travels weekly. (A), (B), and (D) can be determined using strictly the data. Although (C) can be determined by matching up the whale data with weather data, it would require the addition of another data set. Since the question was what can be determined by just using the one data set, the answer is (C). That was a whale of a question.

Large data sets are useful in finding patterns in the data that can predict future usage. Which of the following is used to predict future usage?

• Calculating the monthly bill charged

• Using past purchases to recommend products for possible purchase

• Identifying the largest-spending addresses

• Identifying the time at which most purchases are made

Answer explanation

Companies use computer programs to process information to gain insight and knowledge. Large data sets provide opportunities and challenges for extracting information and knowledge. Past purchases can be used to suggest future purchases. For example, if a dog leash is purchased from Amazon, the next time the customer visits Amazon, the site might suggest dog food. Choices (A), (C), and (D) do not make predictions but, instead, make calculations based on the actual data.

Suppose that a company is creating an algorithm that matches people to advertisements in which they might have an interest. The company is basing its algorithm on a data source accumulated from data bought from social media. Which of the following sources would be useful for the algorithm?

• Information about a person’s video-watching preferences (whether they watch viral videos, reviews, animations, etc.)

• A social network’s information on what sites a person likes and follows

• Information on a person’s previous purchases

• All of the above

Answer explanation

Combining data sources, clustering data, and data classification are all part of gaining insight and knowledge of customer preferences. There is little data that is not filtered and cleaned for use in targeted advertisements. A person’s viewing habits might show interests or where the user is likely to view ads. Which sites a user follows and likes can lead to sales in products related to those sites. Previous purchases like buying a pet would be valuable to pet food companies.

A short message service (SMS) can be sent from one phone to another phone. Once the message is converted into binary form, it is sent by radio waves to the control tower. Once the SMS travels through the control tower, it arrives at the short message service center to be stored or sent immediately. In addition to sending the SMS, the cell phone carrier sends metadata.

Which of the following is NOT metadata?

• The time the message was sent and received

• The content of the SMS

• The geographic location of both the sender and the receiver

• The phone number of both the sender and the receiver

Answer explanation

Metadata are data about the data. The actual data in this example are the content of the SMS message. Metadata are descriptive data and can provide additional information about various aspects of that data. Time (A), geographic location (C), and phone numbers (D) are all metadata.

Spotify is a music-streaming service that uses collected metadata to suggest future playlists. The metadata contain the album name, the genre, the tags describing the music, the mood of the music, and the time and date the music was played.

Which of the following CANNOT be determined using only the information contained in the metadata?

• The most popular current album

• The time when the streaming service is used

• Whether the streaming service is used more during the celebration parade for the Mets winning the World Series

• The least common mood of the audience

Answer explanation

Combining data sources is part of the process of using computers to process information. To determine if the streaming service is used more during a Mets celebration parade, an additional data set that includes weather must be used.

Choices (A), (B), and (D) are all data about the data (metadata). Spotify uses this metadata to suggest music to the user to increase his or her positive experience with Spotify. The goal is for the customer to continue using the product.

An online e-commerce site maintains a database containing the following information:

■Price

■Colors

■Quantity available

■Customer comments from the purchasing public

Using only the database, which of the following CANNOT be determined?

• How some of the buying public feel about their purchases

• The popularity of a color depending on the season

• What color sold the most product

• The average price paid for each color

Answer explanation

Summaries of data analyzed computationally can be effective in communicating insight and knowledge gained from digitally represented information. An additional data set containing date and location would be needed to determine season. (A), (C), and (D) can be determined by examining the data set only.

Google Trends is a website by Google that analyzes the popularity of top search queries in Google Search across various regions and languages. The large number of searches make Google Trends one of the world’s largest real-time data sets. Which of the following is Trends data most likely to answer?

• Whether the public is concerned about the celebration parade for the Jets winning the Superbowl

• What date is the cheapest date to book a hotel room

• The cost of a gas-powered chainsaw

• The length of the Iditarod (a popular sled dog race in Alaska) in meters

Answer explanation

Google Trends looks up what people search but not the content contained on the searched websites. When the public is concerned about a topic, they tend to search to find resources to examine the situation.

(B) Searching for a hotel will not indicate the price of the hotel.

(C) Although people may search for a gas-powered chainsaw, the search itself will not indicate the price.

In 2000, the Chicago Public Schools set up a computing algorithm to detect teachers who had changed their students’ answers on standardized tests. The algorithm looked at students’ answers, with letters representing correct answers, a 0 representing a blank answer, and other numbers representing an incorrect answer. Which of the following is a way that the algorithm could have used this data?

• To look for consecutive answer patterns shared by students in the same class

• To look at the number of correct answers on harder questions

• To look for consecutive correct answers shared by students in the same class

• To look at the percentage of students who passed the test

Answer explanation

The algorithm was set up to look for patterns in the data, whether that was a string of correct answers shared by many students or the same incorrect answer being shared by many students. (For example, in one instance, 15 of 22 students shared a string of seven identical answers. Six of these were correct answers for the hardest questions, with no comparable string in the easier sections.)

(B) and (D) are both too broad and would likely catch teachers with a large number of prepared students.

(C) leaves out the search for incorrect answers, which can be even more telling because students rarely reach the wrong conclusion at the same time.

A large data set contains information on students about to take the College Board’s SAT test, including the following:

■The student’s parents’ highest education level

■The student’s grade point average (GPA)

■The student’s intended college major

■The college that the student is interested in attending

■The student’s current address

■The student’s high school name and location

Which of the following could NOT be answered by analyzing only information in the student data set?

• The number of students accepted to a college

• The number of students interested in a particular college major

• The average GPA for students with an intended college major

• The college with the most interest from a particular high school

Answer explanation

The College Board collects data to gain insight on students taking College Board tests. Although the College Board collects significant data, that data does not include if a student was accepted into and attended a college. The College Board will combine the data to gain insight and knowledge, but it has no access to either student or college decisions on acceptance from this data set