HiGitClass: Keyword-Driven Hierarchical Classification of GitHub Repositories

This repository contains the source code for HiGitClass: Keyword-Driven Hierarchical Classification of GitHub Repositories.

Links

• Installation
• Quick Start
• Data
• Running on New Datasets
• Citation

Installation

For training, a GPU is strongly recommended.

Keras

The code is based on Keras. You can find installation instructions here.

Dependency

The code is written in Python 3.6. The dependencies are summarized in the file requirements.txt. You can install them like this:

pip3 install -r requirements.txt

Quick Start

To reproduce the results in our paper, you need to first download the datasets and the embedding files. The Machine-Learning dataset (ai/) and the Bioinformatics dataset (bio/) can be downloaded here. Then you need to unzip it and put the two folders under the main folder ./. Then the following running script can be used to run the model.

./test.sh

Level-1/Level-2/Overall Micro-F1/Macro-F1 scores will be shown in the last several lines of the output. The classification result can be found under your dataset folder. For example, if you are using the Bioinformatics dataset, the output will be ./bio/out.txt.

Data

Two datasets, Machine-Learning and Bioinformatics, are used in our paper. Besides the "input" version mentioned in the Quick Start section, we also provide the json version, where each line is a json file with user, text (description + README), tags, repository name, and labels. An example is shown below.

{
  "repo": "Natsu6767/DCGAN-PyTorch",
  "user": "Natsu6767",
  "text": "pytorch implementation of dcgan trained on the celeba dataset deep convolutional gan ...",
  "tags": [
    "pytorch",
    "dcgan",
    "gan",
    "implementation",
    "deeplearning",
    "computer-vision",
    "generative-model"
  ],
  "name": [
    "DCGAN",
    "PyTorch"
  ],
  "labels": [
    "$Computer-Vision",
    "$Image-Generation"
  ]
}

NOTE: If you would like to run our code on your own dataset, when you prepare this json file, make sure you list the labels in the top-down order. For example, if the label path of your repository is ROOT-A-B-C, then the "labels" field should be ["A", "B", "C"].

Dataset statistics are as follows.

Running on New Datasets

We use ESim in the embedding module. In the Quick Start section, we include a pretrained embedding file in the downloaded folders. If you would like to retrain the embedding (or you have a new dataset), please follow the steps below.

1. Create a directory named ${dataset} under the main folder (e.g., ./bio).

2. Prepare three files:
   (1) ./${dataset}/label_hier.txt indicating the parent children relationships between classes. The first class of each line is the parent class, followed by all its children classes. The root class must be named as ROOT. Tab is used as the delimiter.
   (2) ./${dataset}/keywords.txt containing class-related keywords for each leaf class. Each line has a class name and a keyword.
   (3) ./${dataset}/${json-name}.json. You can refer to the provided json files for the format. All fields except "repo" are required.

3. Install the dependencies GSL and Eigen. For Eigen, we already provide a zip file ESim/eigen-3.3.3.zip. You can directly unzip it in ESim/. For GSL, you can download it here.

4. ./prep_emb.sh. Make sure you change the dataset/json names.

After that, you can train the classifier as mentioned in Quick Start (i.e., ./test.sh). Please always refer to the example datasets when adapting the code for a new dataset.

Citation

If you find the implementation useful, please cite the following paper:

@inproceedings{zhang2019higitclass,
  title={HiGitClass: Keyword-Driven Hierarchical Classification of GitHub Repositories},
  author={Zhang, Yu and Xu, Frank F. and Li, Sha and Meng, Yu and Wang, Xuan and Li, Qi and Han, Jiawei},
  booktitle={ICDM'19},
  pages={876--885},
  year={2019},
  organization={IEEE}
}