Fast Counting in Machine Learning Applications
This work addresses computational bottlenecks in machine learning for applications like Bayesian networks and association rule mining, though it appears incremental as an improvement over existing methods.
The paper tackles the problem of efficiently executing counting queries in machine learning applications by proposing scalable methods that abstract queries for stream-based aggregation. The result shows that these methods significantly outperform ADtrees and hash tables, offering practical alternatives for large-scale data processing.
We propose scalable methods to execute counting queries in machine learning applications. To achieve memory and computational efficiency, we abstract counting queries and their context such that the counts can be aggregated as a stream. We demonstrate performance and scalability of the resulting approach on random queries, and through extensive experimentation using Bayesian networks learning and association rule mining. Our methods significantly outperform commonly used ADtrees and hash tables, and are practical alternatives for processing large-scale data.