Unit 5 Review

Which of the following is LEAST likely to be an important consideration for a company when

using crowdsourcing to analyze a large data set?

• How to assure the quality of work received from multiple sources

• Whether any data being shared contains private information or if there are any

  intellectual property rights issues

• Will the participants be located geographically close enough to each other to share

  information.

• Which tasks within the analysis require human intervention, and which just need

  computing power.

When we wish to analyze unstructured data the first step is often to develop a framework for the data and organize the data according to this (for example extracting information from an image by using text recognition). Which of the following statements about this process is true?

• The structured data is more useful than the unstructured data.

• The structured data is more useable than the unstructured data.

• All the data from the unstructured format will still be present in the structured format.

• Each source of unstructured data can only be structured according to a single framework

. A football team has an important playoff game approaching, and the coach has employed a statistician to aid in preparations. One of the questions the statistician is seeking to answer is “What offensive strategy should the team use in the second half to maximize their chances of winning if they find themselves down at halftime by 20 points?” Which of the following best describes the branch of statistical analysis which seeks to answer questions like the above?

• Desciptive

• Predictive

• Prescreptive

• Clustering

A state exam board employs a number of external examiners who mark exam papers which need to be manually marked. The board keeps a record of each exam paper which is manually marked containing the following data: ● The title of the exam to which the paper relates ● The name of the student who wrote the paper ● An ID code for the student who wrote the paper ● The grade assigned to that paper by the examiner ● The name of the external examiner who marked the paper ● The date on which the paper was marked A database is created from these records for each season. Which of the following CANNOT be determined from this database?

• How many external examiners did not mark any papers that season.

• The date on which the greatest number of papers were marked

• How many students had more than one paper marked by an external examiner.

• The average grade awarded by each external examiner who marked papers.

Certain programs are designed to analyze electronic scans of documents such as receipts, business cards and recipes, and allow information from these paper copies to be represented electronically. In order for such a program to understand the difference between, for example, a zip code and a phone number or between the price of an item and the total bill for a shopping trip, certain rules must be put into place. Which of the following rules would be most likely to correctly identify the part of a business card which contains an email address? (The character ___ here means a string of any length without spaces).

• Search for any string containing the character “@

• Search for strings with the format “____.com

• Search for strings with the format “___@____.com”

• Search for strings with the format “____@____”

Which of the following best describes the general relationship between confidence and utility in data analytics?

• As utility is increased, confidence will also increase

• As utility is increased, confidence will decrease.

• Utility and confidence are not related to one another.

• As utility is increased confidence may increase or decrease depending on the situation

Which of the following is a technique which could be used to extract structured information from unstructured data?

• Running a regression analysis of the data set

• Creating a database query

• Creating a framework for the information

• Using prescriptive analytics

Which of the following is a risk of obtaining information through the use of crowdsourcing?

• Crowdsourcing requires private information to be publicly shared with competitors.

• Unless independently verified, the results of crowdsourcing may be inaccurate.

• Crowdsourcing can only address small-scale problems, but cannot be used to solve large-scale problems

• The combination of multiple perspectives cannot produce greater insight and knowledge that can be obtained when working alone

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

• The number of purchases made by an individual customer

• The product code of the most-purchased item on the site in a given year.

• The average cost of an individual item purchase on the site.

• How many times a certain product was purchased in a given month.

Which of the following best describes what happens when we take unstructured data and organize it into structured data?

• When we extract the data from an unstructured source and organize it into a structured data set, no data is lost, it is just in a more usable format.

• When data is taken from an unstructured data source and organized into a structured set of data, some data is lost, but the data is now much more useful.

• When unstructured data is organized into structured data there is some loss of data, but the data is in a much more usable format.

• When unstructured data is reorganized into a structured format there is no loss of data, it just simply becomes more useful in the new format.