business intelligence and data analytics

Question 1

What are the benefits of these large data sets?

Answer 1

New, better, cheaper products and services, deep insight into behavior.

Question 2

What are the challenges associated with large data sets?

Answer 2

High complexity of behavior (more detail in the data, dimensionality), traditional analytics and engineering insufficient.

Question 3

What is master data, and what are some examples?

Answer 3

Master data relates to core entities a company works with, such as customers, products, employees, suppliers, and vendors.

Question 4

What is transactional data, and what are some examples?

Answer 4

Transactional data captures timing, quantity, and items related to business activities. Examples include POS data, credit card transactions, money transfers, web visits, and RFM variables (Recency, Frequency, Monetary).

Question 5

What is external data, and what are some examples?

Answer 5

Answer: External data is collected from sources outside the organization. Examples include social media data, macroeconomic data, weather data, competitor data, search data, and web-scraped data.

Question 6

What is open data?

Answer 6

Open data is data that anyone can access, use, and share without copyright restrictions.

Question 7

What are the five V’s of big data?

Answer 7

Volume (data at rest), Velocity (data in motion), Variety (data in its many forms), Veracity (data in doubt), and Value.

Question 8

What is structured data, and what are some examples?

Answer 8

Structured data is organized in a predefined format, like a table. Examples include numbers and the name of a customer.

Question 9

What is unstructured data, and what are some examples?

Answer 9

Unstructured data is not organized in a predefined format. Examples include product reviews and Tweets.

Question 10

What is semi-structured data, and what are some examples?

Answer 10

Semi-structured data has some structure, but it’s not as rigid as structured data. Examples include web pages and resumes.

Question 11

What is metadata?

Answer 11

Metadata is data that describes other data, often referred to as “data about data.

Question 12

What are some examples of metadata?

Answer 12

Data definitions, stored in the catalog of a DBMS. In fraud detection, metadata might include the date and location of an accident picture. In web analytics, metadata could include web visits and geospatial analysis.

Question 13

What kind of information do banks gather to assess credit risk?

Answer 13

Banks gather information about an individual’s default behavior, such as date of birth, gender, income, and employment status.

Question 14

What is up-selling?

Answer 14

Up-selling is offering a customer a higher-priced or premium version of a product they were originally considering. For example, suggesting Westmalle instead of Stella Artois.

Question 15

What is cross-selling?

Answer 15

Cross-selling is offering a complementary product to a customer’s original purchase. For example, pairing cheese with Westmalle.

Question 16

What is down-selling?

Answer 16

Down-selling is offering a customer a lower-priced or less feature-rich version of a product when they can’t afford or don’t need the original choice. For example, suggesting fewer beers if a customer has already ordered too many.

Question 17

What is RFM analysis used for?

Answer 17

RFM is a well-known and well-developed measurement framework used in marketing across various industries. It helps monitor customers’ behavior and develop suitable CRM strategies.

Question 18

What is the focus of RFM analysis?

Answer 18

RFM analysis focuses on existing customers.

Question 19

How does RFM analysis summarize consumer purchasing habits?

Answer 19

RFM summarizes purchasing habits using three dimensions: Recency, Frequency, and Monetary.

Question 20

What is the Pareto principle, and how does it relate to RFM?

Answer 20

The Pareto principle states that 80% of effects come from 20% of causes. In RFM, this means 20% of customers generate 80% of profit.

Question 21

What are the three dimensions of RFM analysis?

Answer 21

Recency (time since most recent purchase), Frequency (total number of purchase transactions), and Monetary (value of purchases).

Question 22

Retention Modeling and Churn Prediction

Answer 22

Retention modeling predicts which customers will leave or decrease their product/service usage.

Question 23

Why is customer retention important?

Answer 23

Long-term loyal customers generate higher profits, are less sensitive to competition, are less costly to serve, have a positive word-of-mouth effect, and have a higher CLV.

Question 24

What are the different types of churn?

Answer 24

Answer: Active churn (customer stops the relationship), passive churn (customer decreases the intensity of the relationship), forced churn (company stops the relationship), and expected churn (customer no longer needs the product/service).

Question 25

Transaction versus Relationship buyers

Answer 25

– transaction buyers: buy because of low price
– relationship buyers: build loyal relationship with firm

Question 26

Goal and Focus of response modelling

Answer 26

Goal: Model response to marketing campaigns.
Focus:
Customer acquisition.
Deepening customer relationships.

Question 27

Targeted Marketing

Answer 27

Targeted Marketing: Delivering marketing messages to specific individuals based on their characteristics and behaviors.
Classification Task: Predicting whether a customer will respond to a marketing campaign (e.g., purchase, click, unsubscribe)

Question 28

Implicit and Explicit Responses

Answer 28

Implicit Response: Actions that indicate interest but may not be direct purchases.
Reading an email
Clicking on a link
Downloading a product description
Configuring a product
Contacting customer service
Explicit Response: Direct actions related to purchasing and advocacy.
Purchase
Purchase and good review
Purchase, good review, and word-of-mouth

Question 29

Definition of Recommender Systems

Answer 29

Recommender Systems: Software tools and techniques that provide suggestions for items to be of use to a user.

Question 30

data of geospatial apps

Answer 30

Historical CRM Dataset: Contains information about customers, including payment history.
Open Data from StatBel: Provides socioeconomic data about different neighborhoods.

Question 31

Approach of geo spatial

Answer 31

Neighborhood Characterization: Use open data to describe neighborhoods based on socioeconomic factors.
Geocoding: Match customer addresses to specific neighborhoods.
Model Development: Create a model to predict the likelihood of a customer becoming a “bad payer” based on neighborhood characteristics and other relevant factors.

Question 32

Customer Lifetime Value (CLTV):

Answer 32

CLTV = (R_t - C_t) * s_t / (1 + d)^t
Where:
R_t: Revenue in month t
C_t: Cost in month t
s_t: Survival probability in month t
d: Discount rate

Question 33

In a subscription or contractual setting churn can be defined

Answer 33

as the customer explicitly cancelling the contract.

Question 34

leave you or decrease their product/service usage.

Answer 34

churn prediction

Question 35

Customer journey analysis can be used to

Answer 35

get a clear and comprehensive picture of the overall process.
highlight process deficiencies such as excessive processing times, indicate deadlock situations, circular references, and unwanted customer leakage, among others.

verify if the process is compliant with both internal and external regulations.

Question 36

RFM analysis is sometimes referred to as

Answer 36

a poor man’s approach to customer lifetime value (CLV) analysis.

Question 37

The RFM framework is a well-known and well-developed measurement framework

Answer 37

The RFM framework is a well-known and well-developed measurement framework used in marketing across different industries such as banking, insurance, Telco, non-profit, travel, on-line retailers, and even government.

Question 38

A web page is an example of

Answer 38

semi-structured data.

Question 39

credit scoring

Answer 39

It is the aim of credit scoring to come up with a statistically based decision model which allows to score future credit applications and decide which ones to accept or reject.

Question 40

In credit scoring, the target variable .

Answer 40

In credit scoring, the target variable is binary.

Question 41

Information retrieval

Answer 41

Information retrieval retrieves documents or web pages based on search terms. An example of this is the Google search engine.

Question 42

Text summarization

Answer 42

Text summarization summarizes text into a few concepts or keywords. This could be useful in a complaint analysis application.

Question 43

Text classification assigns

Answer 43

Text classification assigns text to a set of predefined categories. Think about spam filtering where emails are classified as spam or not based on their context.

Question 44

Search data such as Google Trends can be used for nowcasting where the aim is to

Answer 44

forecast the present or near future.

Question 45

Credit bureaus are

Answer 45

data pooling organizations that gather default information from various financial institutions such as delinquency history, bureau checks, and bureau score.

Question 46

According to Van Vlasselaer and Baesens, which is a key characteristic of fraud?

Answer 46

well-considered
imperceptibly concealed
time-evolving
carefully organized