Minimizing Polarization and Disagreement in the Friedkin–Johnsen Model with Unknown Innate Opinions
Authors: Federico Cinus, Atsushi Miyauchi, Yuko Kuroki, Francesco Bonchi
IJCAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our experiments on various synthetic and real-world datasets show that we can effectively minimize polarization and disagreement even if we have quite limited information about innate opinions. In this section, we asses our framework and the impact of the reconstruction error on solution quality, compared to the ground-truth opinions. We present experiments on 16 networks with up to 1.6 million edges, considering both real opinions and synthetic opinions generated with varying distributions and polarization levels. Dataset statistics are provided in Table 2. |
| Researcher Affiliation | Collaboration | 1Sapienza University, Rome, Italy 2CENTAI Institute, Turin, Italy 3Eurecat, Barcelona, Spain {name.surname}@centai.eu |
| Pseudocode | Yes | Here, we outline the three steps of our proposed framework, with detailed explanations, pseudocodes, and time complexity analyses provided in the Supplementary Material. |
| Open Source Code | Yes | Our code is available at https://github.com/ Federico Cinus/Query-Min PD. |
| Open Datasets | Yes | Dataset Statistics: Referendum 2,479 154,831 Brexit 7,281 530,607 Vax No Vax 11,632 1,599,220 directed/moreno-highschool 70 366 ... The first three datasets contain direct follow networks on X and real opinions. The other networks obtained from KONECT [Kunegis, 2013] are associated with opinions sampled from Gaussian distributions. |
| Dataset Splits | Yes | We use Degree Centrality to select 20% of nodes in each instance for opinion reconstruction. ... The 20% threshold, indicated in black, represents the selected node size used in all other experiments. |
| Hardware Specification | No | The paper does not provide specific hardware details used for running its experiments. |
| Software Dependencies | No | The paper mentions 'CVX' in the context of solving optimization problems, but no specific version number is provided. For the other objectives and constraint, we apply the projection steps in [Cinus et al., 2023]. We compute a local minimum, using the reconstructed opinions as input. ... For the other objectives and constraint, we apply the projection steps in [Cinus et al., 2023]. |
| Experiment Setup | No | Experimental settings are in the Supplementary Material. ... The main text does not provide specific details such as learning rates, batch sizes, or the number of iterations for the GNN training. |