The icws2019 from fagan2888

icws2019's Introduction

Recommending Packages of Multi-Criteria Items to Groups

This is the source code for the model implementations from the paper:

Ceh-Varela, E., & Cao, H. (2019, July). Recommending Packages of Multi-Criteria Items to Groups. In 2019 IEEE International Conference on Web Services (ICWS) (pp. 273-282). IEEE.

Project Structure

datasets (Datasets for the TripAdvisor's catetories)
- Category 1: Restaurant, Category 2: Hotel
model (Source files for different modules)
packages (Placeholder for the final resulting packages)
parameters (Different parameters used by the modules)
- common_parameters.py
temp (Placeholder for auxiliary files generated during execution)
logs (Placeholder for final metrics results used for the paper's figures)

Example of execution

python run.py
each file inside the "model" folder could be run independently in the following order.
- To generate the auxiliary data (these scripts should be run before the models or baselines). An example is the following:
  - python Extract_Biclusters.py -o 1 -x 5 -y 4
  - python CC_Utils.py -o 1
  - python Get_Items_Based_UF.py -o 1 -kitems 20
  - python Get_SIM_i_f.py -o 1
  - python 'Find_UI_items.py -o 1
  - python Get_CC_U-I_Belong.py -o 1
  - python I-I-CF.py -o 1 -G 158,204,41,98
  - python U-U-CF.py -o 1 -G 158,204,41,98
  - pytho Predict_Ratings.py -o 1 -alpha 0.6
  - python Build_Dense_Matrix.py -o 1 -kitems 20 -G 58,204,41,98
- After running the above scripts for all categories (i.e., '-o 1', and '-o 2')
  - python Get_POC.py -o all -G 58,204,41,98
  - python Extract_User_Preferences.py -o all -agg 5 --pt 0
- To run the models
  - python Model1.py -o all -agg 5 --pt 0
  - python Model2.py -o all -agg 5 --pt 0
- To run the baselines
  - python Fairness_baseline -o all --kitems 20
  - python Average_baseline -o all --kitems 20
  - python GRmodel_baseline -o all --kitems 20
- To run the metrics
  - python Consensus_metric.py
  - python m_envy_metric.py

Recommend Projects