h4iku / tag-recom

Data and source code of the paper "A Content-Based Model for Tag Recommendation in Software Information Sites."

How to Run

1. Install Python 3.10.

2. Clone this repository and create a venv:

   git clone https://github.com/h4iku/tag-recom.git
   cd tag-recom
   python -m venv .venv
   .venv/Scripts/activate

3. Update and install the required dependencies:

   python -m pip install -U pip setuptools
   pip install -r requirements.txt

4. Download the datasets from here, and unzip them in the root directory of the cloned repository:

   tag-recom/data
   ├── apple
   ├── askubuntu
   ├── codereview
   ├── dba
   ├── serverfault
   ├── softwareengineering
   ├── stackoverflow
   ├── stats
   ├── superuser
   ├── tex
   └── wordpress
   

5. Run the main module:

   python tag_recommender/main.py

   Change the value of the DATASET variable in datasets.py to choose different datasets. There are also some boolean flags in main.py to control the execution of different parts of the program.