https://github.com/tkdsheep/BIKER-ASE2018

The dataset and source code for paper "API Method Recommendation without Worrying About the Task-API Knowledge Gap"
https://github.com/tkdsheep/BIKER-ASE2018

Last synced: 2 months ago
JSON representation

The dataset and source code for paper "API Method Recommendation without Worrying About the Task-API Knowledge Gap"

• Host: GitHub
• URL: https://github.com/tkdsheep/BIKER-ASE2018
• Owner: tkdsheep
• Created: 2018-07-22T05:12:32.000Z (over 6 years ago)
• Default Branch: master
• Last Pushed: 2018-08-20T04:08:59.000Z (over 6 years ago)
• Last Synced: 2024-08-04T00:11:33.701Z (5 months ago)
• Size: 4.88 KB
• Stars: 18
• Watchers: 1
• Forks: 5
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

• awesome-software-engineering-research - BIKER - API Knowledge Gap", ASE, [Paper](https://dl.acm.org/doi/10.1145/3238147.3238191), including about 400 API retrieval tasks from Stack Overflow. (Data Sets and Benchmarks / API Recommendation)

README

        
# BIKER-ASE2018

The dataset and source code for paper "API Method Recommendation without Worrying About the Task-API Knowledge Gap"

Since the dataset is quite large, I have to upload it using Dropbox. Please download the full package using the following link:

https://www.dropbox.com/s/fr4gdbyfn58ytm8/BIKER.zip?dl=0

The code is based on Python 2.7.12. The required packages are listed below:

numpy 1.13.3

gensim 3.2.0

nltk 3.2.5

In the main package, running test.py will output the evaluation result of BIKER (method-level recommendation) with our dataset. Running test_api_class.py will output the result of class-level recommendation.

The online.py file allows you to input any query online, and output the recommended top-5 API methods, along with the API summary. This file may take 30~60 seconds to preprocess the data (just once) it would print "loading data finished" when it is ready to receive your query.