A non-api member recommendation method based on heuristics and neural networks

A neural network, recommendation method technology, applied in the field of non-API member recommendation, can solve problems such as insufficient sample information and dependence

Inactive Publication Date: 2019-04-12
BEIJING INSTITUTE OF TECHNOLOGYGY
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Although existing methods can recommend API members very well, these methods all rely on rich sample information of the recommended API when making recommendations.
For non-API members, since these members only appear in the current project, the sample information is not rich, so the existing method is not suitable for recommending non-API members

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A non-api member recommendation method based on heuristics and neural networks
  • A non-api member recommendation method based on heuristics and neural networks
  • A non-api member recommendation method based on heuristics and neural networks

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0044] This embodiment elaborates in detail the methods and effects of the non-API member recommendation method based on heuristics and neural networks implemented under 9 open source projects.

[0045] Under the hardware environment shown in Table 1, the open source software shown in Table 2 is trained and predicted.

[0046] Table 1: Hardware environment configuration information table

[0047]

[0048]

[0049] Table 2: Basic information table of open source software

[0050]

[0051] Step A: Extract the non-API members accessed on the right side of the assignment statement and their context information from the open source software shown in Table 2, and use the 9-fold cross-validation method to generate the data training set and test set.

[0052] Among them, 9-fold cross-validation refers to sequentially using one of the nine items as test data and the other eight as training data for cross-validation; for the i-th item G i When performing cross-validation, G ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention relates to a non-API object recommendation method based on heuristics and neural networks, and belongs to the technical field of code completion and code recommendation. According to themethod, all members included in a non-API object declaration type are collected according to non-API member accessing sample sample on the right side of an assignment statement in open source software, the members comprise members obtained by inheritance, then inaccessible members are eliminated according to the relationship between the class where the assignment statement is located and the non-API object declaration type, and the remaining accessible members are all taken as candidates and put into an initial candidate list cdtList for the use of subsequent use. Prediction is conducted on the sample in step one based on three kinds of specific heuristic rules. The neural networks are trained by using information to obtain a filter capable of filtering low-reliability prediction results.When a programmer enters '.' in a non-API sample object on the right side of the assignment statement, non-API members which might access are predicted. According to the method, the probability thatthe number of recommended members under the same data set is correct is obviously higher than existing methods and tools.

Description

technical field [0001] The invention relates to a non-API member recommendation method based on heuristic and neural network, and belongs to the technical field of code completion and code recommendation. Background technique [0002] Code completion refers to the function of IDE (Integrated Development Environment, Integrated Development Environment) to automatically predict the rest of the code when the programmer types some characters. If the code completion function can correctly predict the sentence to be input by the user, it can effectively improve the coding efficiency. Code completion technology is widely used and is one of the 10 most frequently used commands by programmers in Eclipse. [0003] The recommendation of non-API (Application Programming Interface, application programming interface) members (including methods and fields) is a commonly used code recommendation. When a programmer enters "." after a non-API instance object, the IDE tool will automatically...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): G06F8/30G06N3/08
CPCG06F8/30G06N3/08
Inventor 姜林刘辉
Owner BEIJING INSTITUTE OF TECHNOLOGYGY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap