/BIKER-ASE2018

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

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.