Multi-dimensional evaluation recommendation method based on JAVA Doc knowledge graph

Abstract

The invention discloses a multi-dimensional evaluation recommendation method based on a Java Doc knowledge graph, and belongs to the technical field of database recommendation. The method comprises the following steps: establishing a Java class knowledge graph by crawling and analyzing a Java Doc document; analyzing the path relationship between entities according to the relationship between the classes and the outside, establishing a recommendation function to mine the data, and determining a recommended recommendation domain according to the relationship between the classes; selecting a clustering method based on K-means to perform clustering between classes, and taking a clustering result as a complementary set of a recommendation domain based on knowledge graph mining; and performing multi-dimensional quantitative scoring on the classes in the recommendation domain, performing similarity quantitative scoring on the multi-dimensional isomorphic model, and finally returning the quantitative scores of the classes in the recommendation domain to the user. According to the method and the device, a user can contact a new JAVA built-in class and understand the difference between the JAVA built-in classes, so that the purpose of using a more appropriate API function is achieved.

Description

technical field [0001] The invention belongs to the technical field of database recommendation, and in particular relates to a multidimensional evaluation and recommendation method based on a JAVA Doc knowledge graph. Background technique [0002] The Knowledge Graph is the knowledge base that Google uses to power its search engine. Its main feature is that it can express knowledge formally and establish the relational topological structure between knowledge. The basic unit of the knowledge graph is the RDF (Relational Data Forma) triple of entity-relationship-entity. At present, a large number of knowledge graphs have emerged, among which the representative ones are KnowItAll, YAGO, DBpedia, Freebase, NELL, and Probase. Because it stores data as a graph structure, search and other related operations are no longer limited to string matching, but turn to semantics and contextual connections of entities. By performing knowledge reasoning on the knowledge graph, it is possib...

AI Technical Summary

Problems solved by technology

Since the JDK contains hundreds of built-in data structures and interfaces, the relationship between classes and between classes and interfaces is extremely complicated, which makes people have no opportunity and ability to learn the relationship between classes and classes bit by bit. The difference is that many more suitable JAVA built-in classes and data structures have not been discovered and used by users

When people use JAVA built-in classes, most of them choose the JAVA built-in classes they are familiar with, but these JAVA built-in classes are often referenced in inappropriate occasions.

Images