Replication Data for: Formal Models of Opinion Formation and their Application to Real Data: Evidence from Online Social Networks

Replication Data for: “Formal Models of Opinion Formation and their Application to Real Data: Evidence from Online Social Networks” The repository includes three files: Code.ipynb, X_friends.npz, and X_opinions.csv. The first one is a notebook document used by Jupyter Notebook, which stores the Python code to replicate the results from the Manuscript. To open this file one need install Python and Jupyter. The second file contains the matrix A of users’ friendship connections. The matrix representation is the sparse csr-format. The file X_opinions.csv consists of the matrices X ̂ and X ̂_- stacked together along the horizontal axis. The last two files keep the data, which was originally introduced in (Kozitsin et al., 2019) and located on the repository https://doi.org/10.7910/DVN/OUZY74. For convenience of usage, we copy it here. References Kozitsin, I. V., Marchenko, A. M., Goiko, V. L., & Palkin, R. V. (2019). Symmetric Convex Mechanism of Opinion Formation Predicts Directions of Users’ Opinions Trajectories. 1–5.

《观点形成的形式化模型及其在真实数据中的应用：来自在线社交网络的证据》复刻数据集。本仓库包含三个文件：Code.ipynb、X_friends.npz与X_opinions.csv。首个文件为Jupyter笔记本（Jupyter Notebook）文档，存储了复现本文手稿结果所需的Python代码，需安装Python与Jupyter方可打开该文件。第二个文件存储用户好友连接关系矩阵A，该矩阵采用稀疏csr格式。X_opinions.csv文件包含沿水平轴拼接的矩阵X̂与X̂_-。后两个文件为数据集内容，其最初由Kozitsin等人（2019）提出，原存储于仓库https://doi.org/10.7910/DVN/OUZY74。为便于使用，我们将其复刻至本仓库。参考文献：Kozitsin, I. V., Marchenko, A. M., Goiko, V. L., & Palkin, R. V. (2019). 观点形成的对称凸机制可预测用户观点轨迹方向。1–5.