1. What is stream data mining?
A. Mining static datasets stored in a database
B. Mining data that arrives continuously and in real-time
C. Mining data from social media platforms
D. Mining data for historical trends
Answer: B
(Stream data mining focuses on mining data that is continuously generated in real-time, typically from sources like sensors, logs, or social media.)
2. Which of the following is a key challenge in stream data mining?
A. Handling large volumes of static data
B. Storing all the incoming data for later analysis
C. Processing data in real-time with limited memory and computational resources
D. Reducing the dimensionality of non-stream data
Answer: C
(A major challenge in stream data mining is the need to process data in real-time with limited resources, as it is impractical to store all the incoming data.)
3. In stream data mining, what does the “Sliding Window” technique refer to?
A. Storing only the most recent data points in memory
B. Using a fixed-size subset of the incoming stream data for analysis
C. Analyzing data in time intervals based on a window that moves over the data
D. Analyzing the entire data stream regardless of its size
Answer: B
(The sliding window technique involves keeping only a fixed-size window of recent data points in memory and discarding older data, which helps in dealing with the continuous flow of data.)
4. Which of the following algorithms is commonly used in stream data mining?
A. K-means clustering (with a fixed dataset)
B. Decision trees for continuous data
C. Hoeffding trees (for decision tree learning in streams)
D. Principal Component Analysis (PCA) for stream data
Answer: C
(Hoeffding trees are a well-known algorithm designed for learning decision trees in a stream-based setting, where the data is received in a continuous flow.)
5. What is the key difference between stream data mining and traditional data mining?
A. Stream data mining handles a fixed, small dataset while traditional mining handles large datasets.
B. Stream data mining processes data in real-time, while traditional data mining typically works with batch-processed datasets.
C. Stream data mining is used for historical analysis, while traditional mining is used for predictions.
D. There is no difference between the two methods.
Answer: B
(Stream data mining works with data that is continuously arriving in real-time, unlike traditional data mining, which typically operates on static, batch-processed datasets.)
6. Which of the following is a typical application of stream data mining?
A. Forecasting sales trends based on historical data
B. Analyzing sensor data from an IoT network in real-time
C. Categorizing customer demographics
D. Storing large amounts of data for later analysis
Answer: B
(Stream data mining is typically used in real-time applications like analyzing IoT sensor data, where data is constantly generated and needs to be processed as it arrives.)
7. What does the “drift” refer to in stream data mining?
A. A sudden, one-time change in the stream data
B. Gradual changes in the underlying data distribution over time
C. The process of discarding older data from the stream
D. The handling of missing values in stream data
Answer: B
(Drift refers to gradual, subtle changes in the data distribution over time, which is common in stream data and must be accounted for in real-time analysis.)
8. What is a common technique used to reduce the memory requirements in stream data mining?
A. Storing all the data permanently
B. Sampling the stream and only keeping a subset of the data
C. Using neural networks for data compression
D. Replacing real-time analysis with batch analysis
Answer: B
(Sampling the stream allows for efficient memory usage by keeping only a representative subset of the data, thus reducing memory requirements.)
9. In stream data mining, what is concept drift?
A. When the stream data becomes completely random and unstructured
B. When the data stream no longer follows the same patterns and the model must adapt
C. The process of removing outliers from the stream
D. The continuous movement of the data stream in time
Answer: B
(Concept drift refers to changes in the underlying data distribution over time, which requires the model to adapt to new patterns in the stream data.)
10. Which of the following methods is used to handle infinite data streams in stream data mining?
A. Storing all data points for later batch processing
B. Using a time-limited sliding window to store only a subset of the stream
C. Ignoring new data once the model is trained
D. Using fixed-size memory buffers to hold all incoming data
Answer: B
(A sliding window approach is commonly used to handle infinite data streams by retaining only the most recent data points, ensuring the analysis remains manageable.)
11. Which of the following types of models is most suitable for stream data mining?
A. Static models trained on a complete dataset
B. Online learning models that adapt incrementally as new data arrives
C. Batch learning models that process all data at once
D. Neural networks with large fixed memory requirements
Answer: B
(Online learning models are designed for stream data mining because they adapt incrementally as new data arrives, making them well-suited for real-time processing.)
12. What is the role of data summarization in stream data mining?
A. To store all incoming data for future analysis
B. To create a compact representation of the data that can be processed efficiently
C. To remove noise from the stream
D. To reduce the accuracy of models to improve performance
Answer: B
(Data summarization involves creating compact representations of the data (such as sketches or summaries) that capture important features, allowing for efficient analysis in real-time.)
13. Which of the following is an example of stream data?
A. A customer survey form
B. A sequence of temperature readings from a sensor every minute
C. A database of sales transactions for the past year
D. A list of customer names and addresses
Answer: B
(A sequence of temperature readings from a sensor, which is continuously updated in real-time, is a classic example of stream data.)
14. What is incremental learning in the context of stream data mining?
A. Updating the model periodically by processing all past data
B. Learning a new model for each incoming data point
C. Adapting the model incrementally as new data arrives, without retraining on the full dataset
D. Learning from a fixed set of data after a period of time
Answer: C
(Incremental learning involves updating the model as new data arrives, without the need to retrain from scratch on the full dataset, which is essential for stream data mining.)
15. In stream data mining, which of the following is the primary concern regarding model accuracy?
A. The model must be highly accurate when evaluated on historical data
B. The model must provide real-time predictions and adapt to new data patterns over time
C. The model should store all incoming data to improve its accuracy
D. The model should use complex algorithms that take longer to compute
Answer: B
(The primary concern is to maintain real-time predictions and ensure the model adapts to new data patterns as they emerge, rather than being evaluated on historical data alone.)