Artificial Intelligence in Marketing Flashcards

1
Q

What is AI?

A

A science like Math or biology

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

Why important for digital marketing?

A

Several reasons - it is used in many applications in marketing
ex: predicting consumer behavior, content creation, sales forecasting

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

AI vs Machine Learning (ML) vs Deep Learning (DL)

A

AI - like science or math
ML - a system that learns by itself without need for specific programming, improves when you feed it more data
DL - Type of ML, human neural system mimicked to learn something or to solve a problem

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

what is Overfitting & underfitting when it comes to learning?

A

Overfitting; Not enough data + model too complex for the problem

Underfitting: Not enough training + too simple, can`t explain the variance

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

Recommendation systems

A

Used to recommend users similar or relevant products based on their prefeances, uses user-data

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

3 primary components of a recommender system

A

Candidate generating, Scoring, Re-ranking

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

Candidate generating

A

2 ways to filter content
1. Content based filtering
Uses similarity between items to recommend similar items to what user likes. Example watching cute cat videos will lead to recommendations for other animal videos of the same sort

  1. Collaborative filtering
    Similarities between queries and items simultaneously to provide recommendations. User A is similar to user B, user B likes one vid, User A can get that recommended even if he hasn`t watched a similar video
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is the softmax Deep Neural Network?

A

type of artificial neural network (ANN) architecture. Used to assign input data to one of multiple pre-defined classes or categories.

Input is everything like; watch time, watch history, country, language etc.

Output is video candidates with a probability score

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is the goal of YouTube recommendations?

A

Improve watch time

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is so special about TikTok?

A

the first mainstream consumer app where AI IS the product.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is tiktok recommendations based off?

A

“ByteDance”

3 dimensions:
content data
user data
scenario data

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What does Facebook recommendations show?

A

Next item on Facebook or Instagram feed

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Google Ads: optimised targeting

A

Helps you get more conversions within your budget

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Google ads: Performance Planner

A

Forecasting tool that uses Machine Learning to reveal possibilities of your Google Ads campaigns.

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