Data

Question 1

What are data outliers?

Answer 1

Observations that are abnormal and can significantly distort the results

Can be removed from the data set

BUT must be a clear and valid reason, or may be a risk of data manipulation

Question 2

What is big data?

Answer 2

Data sets so large and varied they are beyond the capability of traditional data-processing

Obtained in addition to the traditional management information data

Provide a deeper understanding of customer’s needs

Question 3

The 4Vs of big data?

Answer 3

Volume: scale and amount of information

Velocity: timeliness of the data

Variety: formats, including structured and unstructured data

Veracity: reliability of information, keeping it clean and free from bias, interpretation

Question 4

The purposes of big data in budgeting?

Answer 4

Identify trends and other correlations

Improve forecasting

Improve overall profitability

Question 5

What is data analytics and data mining?

Answer 5

Analytics: process if collecting, organising and analysing data to generate trends and aid decision making

Data mining: sorting through data to identify patterns and relationships, using algorithms

Question 6

Structured v unstructured data?

Answer 6

Structured: data contained within a field in a data record or file

Unstructured: data not easily contained within structured data fields

Question 7

Ads of big data?

Answer 7

Substantial amount of info can be processed

Different sources

Accurate model of future demand

Understand customer’s preferences

Short term and long term decisions

Provides real time information

Question 8

Dis of big data?

Answer 8

Company needs to be seen as trustworthy

Lack of forecasting tools

Infringement of privacy

Security required to hold information

Incorrect data

Lack of skilled data analysis

Question 9

AI versus Machine Leaning?

Answer 9

AI - use of computers to do tasks which are thought to require human intelligence

ML - field in AI whereby computers and learn and do things rather than follow pre-programmed rules

Question 10

The 7 types of data bias?

Answer 10

Selection - sample size doesn’t represent population

Self-selection - individuals select themselves

Observer - researcher allows assumptions to influence the observation

Omitted variable - key data not included

Cognitive bias - presentation of data is misleading

Confirmation - people see data that confirms their beliefs

Survivorship - sample only contains items that survived a previous event