Adaptive-Hypergraph-Learning-with-Multi-Stage-Optimizations-for-Image-and-Tag-Recommendation

This is a Matlab implementation of Hypergraph multi-stage optimization scheme (HMSO) and HMSO using Least Mean Squares (LMS), i.e., HMSO (LMS), of Adaptive Hypergraph Learning with Multi Stage Optimizations for Image and Tag Recommendation journal paper.

run HMSO

In order to run the proposed HMSO approach for the Greek POIs dataset, navigate to folder "experiments_on_greek_pois_dataset" and run the script main.m. All necessary functions to run main.m are included in folder "functions".

HMSO results compared to other methods

run HMSO (LMS)

In order to run the proposed HMSO (LMS), navigate to folder "experiments_on_thessaloniki_dataset" and run the script main.m. All necessary functions to run main.m are included in folder "functions". Note: running HMSO (LMS) on a big dataset, such as greek_pois_dataset, has extremely high time requirements. Thus, we strongly recommend you run HMSO (LMS) on datasets with limited size.

HMSO (LMS) results compared to HMSO

Cite

If you find this code useful in your research, please consider citing:

@article{karantaidis2021,
  title={Adaptive Hypergraph Learning with Multi Stage Optimizations for Image and Tag Recommendation},
  author={Karantaidis G. and Sarridis I. and Kotropoulos C.},
  journal={Signal Processing: Image Communication},
  volume={pending...},
  number={pending...},
  pages={pending...},
  year={2021},
  publisher={Elsevier},
  doi={https://doi.org/10.1016/j.image.2021.116367}
}