Privacy + Data Sci/Machine Learning Basics

Question 1

What is Artificial Intelligence?

Answer 1

enabling computers with intelligence to solve complex problems

e.g.
robots
chatbots
online gaming
voice assistants

Question 2

What is machine learning?

Answer 2

extract knowledge from data to learn from that data and make predictions

E.g. recommendations
search algorithms
classification

Question 3

What is Data Science?

Answer 3

Business and Problem Solving using descriptive and predictive analysis
e.g.
retail trends
financial analysis
transit development

Question 4

According to the Harvard Business Review, what is data science about?

Answer 4

Data science is about infrastructure, testing, using machine learning for decision making, and data
products.

Use data analysis to get insights

Question 5

What can DS work help reveal?

Answer 5

hidden impacts

Question 6

What are the 5 stages of the Data Science Lifecycle?

Answer 6

Problem definition
Data Collection/curation
Data Analysis
Context
Decision Making

Question 7

What questions do we try to answer in problem definition/formulation?

Answer 7

What problem are we trying to solve?
What do we want to find out that we don’t know now?
What are our assumptions and hypotheses
How will we measure success?

Question 8

What are the 2 types of questions we ask in problem definition/formualtion?

Answer 8

Exploratory –> Relationship between different elements/variables/features

Descriptive –> describe statistically, or through data, the situation
Helpful in determining what data you need

Question 9

What questions do we answer in Data Collection and Curation?

Answer 9

What data do we need?
How do we collect that data?
Do we have enough data?
Does the data contain what we are looking for?
What is the scope of the data?
Is the data clean?

Question 10

What do we do in the Data Analysis step?

Answer 10

from simple statistical understanding to complex models for prediction or inference.
- statistical summaries;
- discover patterns;
- machine learning algorithms
- supervised vs unsupervised vs semi-supervised learning.
- Classification - predict a label (discrete)
- Regression - predict a continuous value

Question 11

What kinds of questions do we answer in the data analysis step?

Answer 11

Are we able to answer our question?
Do we have the right form of data and answer?
Are we allowed to use this data?
Are there biases in the data or result?
Can we explain the results?

Question 12

What kind of questions do we answer in the context step?

Answer 12

What world or context are we in?
Does this analysis generalize to other data?

Question 13

What do we do in the context step of the data science lifecycle?

Answer 13

Assess whether inferences carry over from samples to populations
Understand likelihood of predictions with intervals

Analyze Correlation vs Causation

Question 14

What are examples of correlation?

Answer 14

discovery of patterns
passive data collection

Question 15

What are examples of causation?

Answer 15

randomized experiments as needed
active data - interactive data collection

Question 16

What do we do in the decision making step of the data science lifecycle?

Answer 16

What is the data telling us?
Does it answer the right question?
Do we trust our conclusions and decisions?
Will this be valid tomorrow?
How to act on the analysis?

Question 17

Where should we consider privacy?

Answer 17

Input : data collection/sharing
Analysis: how data handled, shared, algorithms
Output: how analysis is presented, how models are published

Question 18

What are the types of information disclosure?

Answer 18

Attribute disclosure
Identity Disclosure
Membership Disclosure

Question 19

What is Attribute Disclosure?

Answer 19

Disclosure of some information about a known person

e.g. healthcare data: diagnosis: test results, etc

Disclosure is through linkage with other data

Question 20

What is identity disclosure?

Answer 20

Identification of a person
De-anonymization
Re-identification
Reconstruction of dataset

Question 21

What is membership disclosure?

Answer 21

Membership in a dataset revealed
Identity and attribute disclosure

Question 22

What is anonymization?

Answer 22

A common tool used to claim privacy, especially in data sharing.
Personal Identifiable Information (PII) is masked or hidden.

Question 23

What is a PII?

Answer 23

Personally Identifiable information
personal information includes any factual or subjective
information, recorded or not, about an identifiable individual.

age, name, ID numbers, income, ethnic origin, or blood type;

Question 24

What is a dataset?

Answer 24

database: the form in which data is presented.
a table, “matrix”.
rows correspond to data points (depending on the data, a row is about an
individuals if the data is about people, or an event for sensor data, or a transaction
for bank data).
columns correspond to features or attributes

Question 25

What is the target?

Answer 25

feature/attribute we are trying to predict

Question 26

How can demographics uniquely identify individuals?

Answer 26

A single attribute: if the frequency of a particular value of an attribute is low. More than one attribute: combinations of attributes can combine to occur even less frequently

Question 27

What did Sweeney's experiment show?

Answer 27

re-identification of individuals is possible when only a single dataset is shared, and when multiple datasets where one is anonymized are shared.

Question 28

What is anonymous data according to sweeney?

Answer 28

- a record by itself cannot be linked to an individual;

Question 29

what is an explicit identifier according to sweeney?

Answer 29

with no additional information the person can be directly found ({name, address} or {name, phone});

Question 30

What is de-identified data according to sweeney?

Answer 30

all explicit identifiers removed, generalized, or replaced (name, address, phone);

Question 31

What is a quasi-identifier according to sweeney?

Answer 31

a set of data elements that are not explicit identifiers that in combination associates uniquely or almost uniquely to an individual

Question 32

How do linkage attacks work?

Answer 32

If an anonymized dataset is released publicly, through the notion of unique combinations and linking with another public dataset with identifiable information, we can re-identify individuals.

Question 33

what is direct linking?

Answer 33

direct linking refers to the process of identifying an individual by correlating anonymized data with another dataset that contains identifying information. relies on explicit attributes

Question 34

What is linking through similarity?

Answer 34

Linking through similarity refers to the process of re-identifying individuals in an anonymized dataset by correlating patterns or attributes with another dataset that contains identifiable information. relies on statistical similarity

Question 35

What is k-anonymity?

Answer 35

The information (identifier or quasi-identifier) contained for each individual in the released dataset cannot be distinguished from at least k − 1 individuals whose information is also in the released dataset. Any quasi-identifier present in the released table must appear in at least k records.

Question 36

Why can't k-anonymity be guaranteed?

Answer 36

Real world datasets are very sparse if project into low dim, we lose info these datasets provide low utility independent releases can be linked to infer info difficult to achieve (NP-hard)

Question 37

What are the goals of k-anonymity?

Answer 37

membership disclosure is protected sensitive attribute protected identity disclosure protected

Question 38

What is a homogeneity attack?

Answer 38

k-anonymity can create groups that leak information due to lack of diversity in the sensitive attribute.

Question 39

What is a background knowledge attack?

Answer 39

k-anonymity does not protect against attacks based on background knowledge

Question 40

What is l-diversity

Answer 40

Let a q∗-block be a set of tuples such that its non-sensitive values generalize to q∗. A q∗-block is ℓ-diverse if it contains ℓ "well represented" values for the sensitive attribute S. A table is ℓ-diverse, if every q∗-block in it is ℓ-diverse. An equivalence class is said to have ℓ-diversity if there are at least ℓ well-represented values for the sensitive attribute. A table is said to have ℓ-diversity if every equivalence class of the table has ℓ-diversity

Question 41

What are the limitations of l-diversity?

Answer 41

May be difficult and unnecessary to achieve insufficient to prevent attribute disclosure, as shown with the following two potential attacks: similarity attack and skewness attack

Question 42

t-closeness

Answer 42

An equivalence class is said to have t-closeness if the distance between the distribution of a sensitive attribute in this class and the distribution of the attribute in the whole table is no more than a threshold t. A table is said to have t-closeness if all equivalence classes have t-closeness.