Q.1 What is the main concern of data mining security?
Improving computational efficiency
Protecting sensitive information from unauthorized access
Increasing data storage
Enhancing graphical representation of data
Explanation - Data mining security focuses on ensuring that sensitive data is not exposed to unauthorized users during data mining processes.
Correct answer is: Protecting sensitive information from unauthorized access
Q.2 Which of the following is a privacy-preserving data mining technique?
Clustering
Data masking
Decision trees
Association rule mining
Explanation - Data masking hides sensitive information by replacing it with fictitious but realistic values, helping preserve privacy during mining.
Correct answer is: Data masking
Q.3 What does 'k-anonymity' in data privacy ensure?
Every record is indistinguishable from at least k-1 others
Every record has a unique identifier
Data is encrypted with k-bit key
k number of users can access the data
Explanation - K-anonymity ensures that individual records cannot be distinguished from at least k-1 other records, preventing re-identification.
Correct answer is: Every record is indistinguishable from at least k-1 others
Q.4 Which type of attack exploits knowledge of data mining models to extract sensitive data?
Model inversion attack
SQL injection
Denial of Service
Phishing
Explanation - In a model inversion attack, attackers use the outputs of a trained model to infer sensitive information about the training data.
Correct answer is: Model inversion attack
Q.5 What is differential privacy?
A technique to encrypt data for security
A method to add noise to query results to protect individual data
A clustering algorithm
A method to compress datasets
Explanation - Differential privacy adds controlled noise to query results so that individual contributions cannot be distinguished, preserving privacy.
Correct answer is: A method to add noise to query results to protect individual data
Q.6 Which of the following can help prevent inference attacks in data mining?
Data anonymization
Increasing RAM
Faster processors
Data compression
Explanation - Data anonymization removes or masks identifying information to prevent attackers from deducing sensitive data via inference attacks.
Correct answer is: Data anonymization
Q.7 What is a common security concern in distributed data mining?
Ensuring the mining algorithm is parallelizable
Preventing data leakage between nodes
Reducing network latency
Improving storage space
Explanation - In distributed mining, multiple nodes may process data; ensuring that sensitive data is not leaked between nodes is a key security concern.
Correct answer is: Preventing data leakage between nodes
Q.8 Which cryptographic method can be used for secure multiparty computation in data mining?
RSA
Homomorphic encryption
MD5 hashing
Base64 encoding
Explanation - Homomorphic encryption allows computations on encrypted data, enabling secure multiparty computation without revealing raw data.
Correct answer is: Homomorphic encryption
Q.9 What does 'l-diversity' aim to improve over k-anonymity?
Ensures diversity of sensitive values within anonymized groups
Reduces computational complexity
Increases dataset size
Improves model accuracy
Explanation - L-diversity ensures that anonymized groups have diverse sensitive attributes, preventing attacks that exploit homogeneity in k-anonymity.
Correct answer is: Ensures diversity of sensitive values within anonymized groups
Q.10 Which of the following is an attack specifically targeting privacy in data mining?
Data poisoning attack
Re-identification attack
Man-in-the-middle attack
Buffer overflow attack
Explanation - Re-identification attacks try to match anonymized records with external datasets to uncover the identity of individuals.
Correct answer is: Re-identification attack
Q.11 Which approach can be used to secure sensitive data in data warehouses?
Access control and encryption
Faster query processing
Data normalization
Indexing
Explanation - Securing data warehouses involves controlling who can access data (access control) and encrypting stored data to prevent unauthorized access.
Correct answer is: Access control and encryption
Q.12 Which of these is a challenge in privacy-preserving data mining?
Maintaining data utility while preserving privacy
Increasing data redundancy
Optimizing network speed
Designing user interfaces
Explanation - Privacy-preserving data mining must balance between hiding sensitive information and keeping the data useful for mining tasks.
Correct answer is: Maintaining data utility while preserving privacy
Q.13 What is a 'data poisoning attack' in the context of secure data mining?
Adding malicious data to affect mining outcomes
Removing all sensitive data
Encrypting all data before mining
Querying data multiple times
Explanation - Data poisoning attacks introduce incorrect or malicious data to manipulate the results of data mining algorithms.
Correct answer is: Adding malicious data to affect mining outcomes
Q.14 Which type of encryption allows querying on encrypted data without decrypting it?
Symmetric encryption
Asymmetric encryption
Homomorphic encryption
Hashing
Explanation - Homomorphic encryption permits computations directly on encrypted data, enabling privacy-preserving queries.
Correct answer is: Homomorphic encryption
Q.15 Which technique reduces the risk of sensitive attribute disclosure while sharing data?
Data anonymization
Data indexing
Data clustering
Data sorting
Explanation - Anonymization removes identifiers or perturbs data to prevent disclosure of sensitive attributes when sharing datasets.
Correct answer is: Data anonymization
Q.16 What is an 'inference attack' in data mining privacy?
Using partial data to deduce sensitive information
Crashing the mining system
Stealing encryption keys
Deleting records from the dataset
Explanation - Inference attacks attempt to infer sensitive data from seemingly innocuous information in the dataset or model outputs.
Correct answer is: Using partial data to deduce sensitive information
Q.17 Which of the following protects privacy by grouping similar records together?
Clustering-based anonymization
Decision tree learning
Frequent itemset mining
Data compression
Explanation - Clustering-based anonymization groups similar records and masks identifying information to prevent disclosure of individual data.
Correct answer is: Clustering-based anonymization
Q.18 Which of these is a limitation of k-anonymity?
Vulnerable to homogeneity attacks
Cannot be applied to large datasets
Requires symmetric encryption
Slows down clustering algorithms
Explanation - K-anonymity does not prevent sensitive attributes in a group from being identical, allowing attackers to deduce sensitive information (homogeneity attack).
Correct answer is: Vulnerable to homogeneity attacks
Q.19 Which security measure ensures that only authorized users can access data mining outputs?
Access control
Data normalization
Data compression
Clustering
Explanation - Access control mechanisms enforce permissions, ensuring that only authorized users can view or use data mining results.
Correct answer is: Access control
Q.20 How does adding noise to data help in privacy-preserving data mining?
It prevents exact disclosure of individual records
It improves algorithm speed
It reduces data size
It increases memory usage
Explanation - Adding noise perturbs the data values slightly, making it difficult to identify individual records while allowing general patterns to be mined.
Correct answer is: It prevents exact disclosure of individual records
Q.21 Which type of attack attempts to reconstruct sensitive data from anonymized datasets?
Reconstruction attack
SQL injection
Denial-of-service attack
Phishing
Explanation - Reconstruction attacks use anonymized or perturbed data along with background knowledge to reconstruct sensitive individual data.
Correct answer is: Reconstruction attack
Q.22 In secure data mining, which principle aims to ensure minimal data exposure?
Principle of least privilege
Clustering principle
Normalization principle
Indexing principle
Explanation - The principle of least privilege ensures users access only the data necessary for their tasks, reducing potential security risks.
Correct answer is: Principle of least privilege
Q.23 Which technique ensures privacy by splitting data across multiple servers?
Data fragmentation
Data clustering
Data normalization
Data mining
Explanation - Data fragmentation splits sensitive data across multiple servers, reducing the risk that a single compromised server reveals the entire dataset.
Correct answer is: Data fragmentation
Q.24 Which approach combines cryptography and data mining to preserve privacy?
Secure multiparty computation
Frequent itemset mining
Decision tree pruning
Data indexing
Explanation - Secure multiparty computation allows multiple parties to jointly compute data mining functions without revealing their individual data, combining cryptography and mining.
Correct answer is: Secure multiparty computation
Q.25 Which concept ensures that results of a query do not compromise individual privacy?
Differential privacy
Data normalization
Frequent pattern mining
Data compression
Explanation - Differential privacy ensures that adding or removing a single record does not significantly affect query results, protecting individual privacy.
Correct answer is: Differential privacy