5 results about "Item set mining" patented technology

This invention discloses a method, system, and storage medium for frequent itemset mining. It requires only a single scan of the transaction database, eliminating the need to sort frequent itemsets in descending order of their support counts or to reorder the items in the transaction database. Even after data is added to the database, there's no need to restart; instead, new transaction items are directly inserted into the existing pattern heap using the algorithm proposed in this invention, incrementally constructing a new pattern heap before rule mining. The invention includes: acquiring a database containing at least one transaction, where each transaction contains at least one data item; scanning the database and dynamically generating a database pattern heap based on the transaction's data items and a preset pattern heap; recursively pruning the heap nodes of the database pattern heap based on a preset minimum support to obtain a frequent pattern heap; and determining frequent itemsets based on the frequent pattern heap mining.

This invention discloses a method for accurate frequent itemset mining of streaming data, relating to the field of data mining technology, including the following steps: S1, accessing and managing structured streaming data through a distributed streaming processing engine; S2, converting streaming transaction data into a vertical format of item and transaction ID sets, and maintaining the vertical database state using state management functions; S3, filtering infrequent items based on a minimum support threshold, performing equivalence class partitioning, and partitioning the conditional database to each computing node based on the equivalence classes; This invention, by deeply integrating an incremental mining architecture with a vertical data format with a distributed streaming processing engine, ensures the complete accuracy of frequent itemset mining results while achieving a high throughput of tens of thousands of transactions per second, thus successfully breaking through the technical bottleneck of "difficulty in balancing accuracy and efficiency" in existing technologies, and achieving significant technical progress.

The invention discloses a software testing method and device, equipment, a storage medium and a product, and relates to the technical field of tests.The software testing method comprises the steps that based on a frequent item set mining algorithm, an interface incidence relation graph is determined according to interface access data; based on a clustering analysis method, determining a service distribution graph according to the key service data; the interface incidence relation graph and the service distribution graph are fused through a graph neural network to obtain a target graph, an intelligent testing scheme is generated according to the target graph, and the graph neural network comprises a two-channel feature coding layer, a cross-modal feature fusion layer, a space-time graph convolution layer and a risk propagation predictor. The test quality can be effectively improved.

The present invention proposes a Spark-based parallel frequent item set mining method, comprising the following steps: S1, obtaining a support count matrix through matrix storage and providing a support count query for creating a conditional FP-tree; then, based on the obtained support count, eliminating items that do not meet the minimum support to obtain a frequent 1-item set matrix; then, decomposing the matrix storing the frequent 1-item sets using a non-negative matrix factorization strategy (SNMF); S2, obtaining a frequent 1-item set F-list from the frequent 1-item set matrix of S1, and uniformly grouping the frequent 1-item sets obtained by S1 using a genetic algorithm grouping strategy (GS-GA); S3, merging conditional pattern bases using an efficient tree structure reduction strategy (ERTS), deleting paths in the FP-tree that do not meet the maximum frequent item set; then, mining the grouped FP-trees in parallel to obtain frequent item sets, and outputting the results to a distributed file system (HDFS). The present invention solves the problems of low spatiotemporal efficiency, high inter-node communication overhead, and redundant search in creating a conditional FP-tree, and significantly improves both spatiotemporal efficiency and recommendation non-empty rate.

The application discloses a legality verification method, device and equipment of a database query statement and a medium. The application decouples the features of historical query statements in a database into two levels, respectively performs frequent item set mining, obtains a first candidate item set and a second candidate item set, constructs an association rule library according to the two candidate item sets, performs matching inspection on a target transaction feature vector of a target query statement through the association rule library, and verifies the legality of the query, thereby realizing online monitoring of potential leakage risks of database access, which is fast, accurate and interpretable.