Week 4 Flashcards

1
Q

What are the data biases ?

A

Confirmation Bias
Selection Bias
Historical Bias
Survivorship Bias
Availability/Recency Bias
Outlier Bias

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What is selection bias ?

A

Selection bias happens when you are using samples that are not representative of the population. This happens when the data selection is not randomised. An example of a selection bias could be if you are asked to predict future sales for a Spanish hotel and you take the sales for July and August and you extrapolate you would have a too high estimate.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is conformation bias ?

A

This is when we search out for confirmation of our own beliefs. This can be subconscious for how we search out information or interpret it. We can do this by focusing on information that confirms our own arguments. Social media is an excellent example

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is historical bias ?

A

Historical data bias occurs when socio-cultural prejudices and beliefs are mirrored in a systematic process. This becomes particularly challenging when data from historically-biased sources are used to train machine learning models—for example, if manual systems give certain groups of people poor credit ratings, and you’re using that data to train the automatic system, the automatic system will replicate and may amplify the original system’s biases.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What is survivorship bias ?

A

Survivorship Bias is the issue of only studying the survivors of a sample. So, if I wanted to know what makes a company successful, I might want to study successful companies like Apple and see what has made them successful. However, I am only looking at companies that have survived and not the ones that have failed.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Availability bias ?

A

This is when you put too much weight on something that has happened recently. An example is swimming in the sea after you hear about shark attacks. If you have recently heard about it then you are going to factor that into your decision-making more than you would have done if you have not heard about it.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Outlier bias ?

A

Outlier bias is when you take an average and ignore the population distribution. So, if we took one of our classes for example and wanted to work the average age. In general, we would expect it to be around 18, however, if you take into account your 80-year-old tutor it would increase the average quite considerably and not paint an accurate picture.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly