diff --git a/Rcodes/simulation/data/simulated_collection_clustering_iid_09-05-23-18:36:54.Rds b/Rcodes/simulation/data/simulated_collection_clustering_iid_09-05-23-18:36:54.Rds deleted file mode 100644 index c460813..0000000 Binary files a/Rcodes/simulation/data/simulated_collection_clustering_iid_09-05-23-18:36:54.Rds and /dev/null differ diff --git a/Rcodes/simulation/data/simulated_collection_clusteringiid-09-05-23-12:00:58.Rds b/Rcodes/simulation/data/simulated_collection_clusteringiid-09-05-23-12:00:58.Rds deleted file mode 100644 index e2ebffa..0000000 Binary files a/Rcodes/simulation/data/simulated_collection_clusteringiid-09-05-23-12:00:58.Rds and /dev/null differ diff --git a/Rcodes/simulation/img/plot_model_function_eps.png b/Rcodes/simulation/img/plot_model_function_eps.png index ba1343d..6124a75 100644 Binary files a/Rcodes/simulation/img/plot_model_function_eps.png and b/Rcodes/simulation/img/plot_model_function_eps.png differ diff --git a/Rcodes/simulation/inference_analyze.tex b/Rcodes/simulation/inference_analyze.tex index da0f782..58e83fe 100644 --- a/Rcodes/simulation/inference_analyze.tex +++ b/Rcodes/simulation/inference_analyze.tex @@ -44,7 +44,8 @@ their connectivity structures encompass a mix of core-periphery, assortative community and disassortative community structures, depending on which 3 of the 4 blocks are selected for each network. \(\eps[\alpha]\) represents the strength of these structures, the -larger, the easier it is to tell apart one block from another. +larger, the easier it is to tell apart one block from another. The true +model of all the simulation is a \(\pi\rho\text{-}colBiSBM\). \paragraph{Inference} @@ -91,9 +92,10 @@ results are provided in the tables \ref{tab:per_model_sep} to \ref{tab:per_model_pirho}. Each line corresponds to the 108 datasets for a given value of value of \(\eps[\alpha]\). +\tiny \begin{table}[!h] -\caption{\label{tab:per_model_table}\label{tab:per_model_sep}Quality metrics for sep-colBiSBM} +\caption{\label{tab:per_model_table}\label{tab:per_model_sep}Quality metrics for $sep\text{-}BiSBM$} \centering \begin{tabular}[t]{rllll} \toprule @@ -114,7 +116,7 @@ $\eps[\alpha]$ & $\overline{\text{ARI}}_{1}$ & $\overline{\text{ARI}}_{2}$ & $\t \end{table} \begin{table}[!h] -\caption{\label{tab:per_model_table}\label{tab:per_model_iid}Quality metrics for iid-colBiSBM} +\caption{\label{tab:per_model_table}\label{tab:per_model_iid}Quality metrics for $iid$$\text{-}colBiSBM$} \centering \begin{tabular}[t]{rllllll} \toprule @@ -135,7 +137,7 @@ $\eps[\alpha]$ & $\overline{\text{ARI}}_{1}$ & $\overline{\text{ARI}}_{2}$ & $\t \end{table} \begin{table}[!h] -\caption{\label{tab:per_model_table}\label{tab:per_model_pi}Quality metrics for $\pi$-colBiSBM} +\caption{\label{tab:per_model_table}\label{tab:per_model_pi}Quality metrics for $\pi$$\text{-}colBiSBM$} \centering \begin{tabular}[t]{rllllll} \toprule @@ -156,7 +158,7 @@ $\eps[\alpha]$ & $\overline{\text{ARI}}_{1}$ & $\overline{\text{ARI}}_{2}$ & $\t \end{table} \begin{table}[!h] -\caption{\label{tab:per_model_table}\label{tab:per_model_rho}Quality metrics for $\rho$-colBiSBM} +\caption{\label{tab:per_model_table}\label{tab:per_model_rho}Quality metrics for $\rho$$\text{-}colBiSBM$} \centering \begin{tabular}[t]{rllllll} \toprule @@ -177,7 +179,7 @@ $\eps[\alpha]$ & $\overline{\text{ARI}}_{1}$ & $\overline{\text{ARI}}_{2}$ & $\t \end{table} \begin{table}[!h] -\caption{\label{tab:per_model_table}\label{tab:per_model_pirho}Quality metrics for $\pi\rho$-colBiSBM} +\caption{\label{tab:per_model_table}\label{tab:per_model_pirho}Quality metrics for $\pi\rho$$\text{-}colBiSBM$} \centering \begin{tabular}[t]{rllllll} \toprule @@ -196,6 +198,36 @@ $\eps[\alpha]$ & $\overline{\text{ARI}}_{1}$ & $\overline{\text{ARI}}_{2}$ & $\t \bottomrule \end{tabular} \end{table} +\normalsize + +\begin{table}[!h] + +\caption{\label{tab:proportion-preferred_model}\label{tab:proportion-preferred-table}Proportions of models selected per \eps[\alpha] (data for Figure \ref{fig:inference-proportion-preferred})} +\centering +\begin{tabular}[t]{rccccc} +\toprule +\eps[\alpha] & $sep\text{-}BiSBM$ & $iid\text{-}colBiSBM$ & $\pi\text{-}colBiSBM$ & $\rho\text{-}colBiSBM$ & $\pi\rho\text{-}colBiSBM$\\ +\midrule +0.00 & 1.00 & 0.00 & 0.00 & 0.00 & 0.00\\ +0.03 & 0.95 & 0.04 & 0.01 & 0.00 & 0.00\\ +0.06 & 0.39 & 0.33 & 0.06 & 0.15 & 0.06\\ +0.09 & 0.07 & 0.01 & 0.12 & 0.25 & 0.55\\ +0.12 & 0.00 & 0.08 & 0.06 & 0.13 & 0.72\\ +\addlinespace +0.15 & 0.00 & 0.12 & 0.08 & 0.08 & 0.71\\ +0.18 & 0.00 & 0.11 & 0.04 & 0.06 & 0.79\\ +0.21 & 0.00 & 0.19 & 0.04 & 0.07 & 0.69\\ +0.24 & 0.00 & 0.09 & 0.06 & 0.08 & 0.77\\ +\bottomrule +\end{tabular} +\end{table} + +\begin{figure} +\centering +\includegraphics{./img/54eb0a21b143a53b6199a869d7a228ad7d158e57.png} +\caption{\label{fig:inference-proportion-preferred}Plot of the +proportions of different preferred models in function of \eps[\alpha]} +\end{figure} \paragraph{Results} @@ -203,3 +235,14 @@ For the model comparison, when \(\eps[\alpha]\) is small (\(\eps[\alpha]\in[0, .04]\)), the simulation model is close to the Erd\H{o}s-Reńyi network and it is very hard to find any structure beyond the one of a single block on each dimension. + +On the figure \ref{fig:inference-proportion-preferred} and table +\ref{tab:proportion-preferred-table} we can see that from +\(\eps[\alpha] = 0.12\) around \(70\%\) of the time the +\(\pi\rho\text{-}colBiSBM\) model (i.e., the correct one) is selected. + +An interesting result we can read in the tables is that our models +outperform the \(sep\text{-}BiSBM\) when considering the ARI on the +whole set of nodes (\(\text{ARI}_d\)). This means that our models are +able to recover the block pairing \emph{between the networks} in +addition to recovering the blocks and their parameters. diff --git a/Rcodes/simulation/model_selection_analyze.tex b/Rcodes/simulation/model_selection_analyze.tex index 754b600..48e8c69 100644 --- a/Rcodes/simulation/model_selection_analyze.tex +++ b/Rcodes/simulation/model_selection_analyze.tex @@ -36,7 +36,7 @@ value of \(\eps[\pi]\) and \(\eps[\rho]\). \(\pi\rho\text{-}colBiSBM\), \(\pi\text{-}colBiSBM\), \(\rho\text{-}colBiSBM\), \(iid\text{-}colBiSBM\) and -\(sep\text{-}colBiSBM\) are put in competition and the model with the +\(sep\text{-}BiSBM\) are put in competition and the model with the greater BIC-L is selected as the \emph{preferred model}. When \(\eps[\pi] = 0\), \(\bm{\pi}^1 = \bm{\pi}^2\), \(\eps[\rho] = 0\) @@ -55,20 +55,20 @@ is a \(\rho\text{-}colBiSBM\). Finally, when \(\eps[\pi] > 0\) or \centering \begin{tabular}[t]{lccccl} \toprule -\multicolumn{1}{c}{ } & \multicolumn{4}{c}{Proportions of model selection} & \multicolumn{1}{c}{Blocks} \\ +\multicolumn{1}{c}{ } & \multicolumn{4}{c}{Models} & \multicolumn{1}{c}{Blocks} \\ \cmidrule(l{3pt}r{3pt}){2-5} \cmidrule(l{3pt}r{3pt}){6-6} $\eps[\pi]$ & $iid\text{-}colBiSBM$ & $\pi\text{-}colBiSBM$ & $\rho\text{-}colBiSBM$ & $\pi\rho\text{-}colBiSBM$ & Recovered $Q_1$\\ \midrule -0.00 & 0.65 & 0.00 & 0.35 & 0.00 & 3.00\\ -0.04 & 0.66 & 0.00 & 0.34 & 0.00 & 3.00\\ -0.07 & 0.64 & 0.01 & 0.34 & 0.01 & 3.01\\ -0.11 & 0.63 & 0.03 & 0.31 & 0.03 & 3.01\\ -0.14 & 0.55 & 0.12 & 0.28 & 0.05 & 3.00\\ +0.00 & 0.65 & 0.00 & 0.35 & 0.00 & 3\\ +0.04 & 0.66 & 0.00 & 0.34 & 0.00 & 3\\ +0.07 & 0.64 & 0.01 & 0.34 & 0.01 & 3.01 $\pm$ 0.01\\ +0.11 & 0.63 & 0.03 & 0.31 & 0.03 & 3.01 $\pm$ 0.01\\ +0.14 & 0.55 & 0.12 & 0.28 & 0.05 & 3\\ \addlinespace 0.18 & 0.39 & 0.26 & 0.21 & 0.13 & 3.01\\ 0.21 & 0.23 & 0.42 & 0.13 & 0.23 & 3.01\\ -0.25 & 0.10 & 0.56 & 0.05 & 0.29 & 3.02\\ -0.28 & 0.01 & 0.65 & 0.01 & 0.33 & 3.01\\ +0.25 & 0.10 & 0.56 & 0.05 & 0.29 & 3.02 $\pm$ 0.01\\ +0.28 & 0.01 & 0.65 & 0.01 & 0.33 & 3.01 $\pm$ 0.01\\ \bottomrule \end{tabular} \end{table} @@ -79,20 +79,20 @@ $\eps[\pi]$ & $iid\text{-}colBiSBM$ & $\pi\text{-}colBiSBM$ & $\rho\text{-}c \centering \begin{tabular}[t]{lccccl} \toprule -\multicolumn{1}{c}{ } & \multicolumn{4}{c}{Proportions of model selection} & \multicolumn{1}{c}{Blocks} \\ +\multicolumn{1}{c}{ } & \multicolumn{4}{c}{Models} & \multicolumn{1}{c}{Blocks} \\ \cmidrule(l{3pt}r{3pt}){2-5} \cmidrule(l{3pt}r{3pt}){6-6} $\eps[\rho]$ & $iid\text{-}colBiSBM$ & $\pi\text{-}colBiSBM$ & $\rho\text{-}colBiSBM$ & $\pi\rho\text{-}colBiSBM$ & Recovered $Q_2$\\ \midrule -0.00 & 0.63 & 0.37 & 0.00 & 0.00 & 3.00\\ -0.04 & 0.65 & 0.34 & 0.00 & 0.01 & 3.00\\ -0.07 & 0.64 & 0.33 & 0.01 & 0.01 & 3.00\\ -0.11 & 0.64 & 0.31 & 0.03 & 0.02 & 3.00\\ -0.14 & 0.53 & 0.29 & 0.11 & 0.06 & 3.00\\ +0.00 & 0.63 & 0.37 & 0.00 & 0.00 & 3\\ +0.04 & 0.65 & 0.34 & 0.00 & 0.01 & 3\\ +0.07 & 0.64 & 0.33 & 0.01 & 0.01 & 3\\ +0.11 & 0.64 & 0.31 & 0.03 & 0.02 & 3\\ +0.14 & 0.53 & 0.29 & 0.11 & 0.06 & 3\\ \addlinespace 0.18 & 0.42 & 0.20 & 0.24 & 0.14 & 3.01\\ 0.21 & 0.25 & 0.12 & 0.40 & 0.22 & 3.01\\ 0.25 & 0.08 & 0.06 & 0.58 & 0.29 & 3.01\\ -0.28 & 0.01 & 0.01 & 0.65 & 0.32 & 3.00\\ +0.28 & 0.01 & 0.01 & 0.65 & 0.32 & 3\\ \bottomrule \end{tabular} \end{table} diff --git a/Rcodes/simulation/netclustering_analyze.tex b/Rcodes/simulation/netclustering_analyze.tex index 290de08..b3e5948 100644 --- a/Rcodes/simulation/netclustering_analyze.tex +++ b/Rcodes/simulation/netclustering_analyze.tex @@ -1 +1,13 @@ \section{Network clustering of simulated networks}\label{sec:network-clustering-of-simulated-networks} + +\begin{verbatim} +## character(0) +\end{verbatim} + +\begin{verbatim} +## list() +\end{verbatim} + +\begin{verbatim} +## Error in eval(expr, envir, enclos): objet 'list_best_partition' introuvable +\end{verbatim} diff --git a/figure/proportion-preferred_model-1.png b/figure/proportion-preferred_model-1.png new file mode 100644 index 0000000..7b48cf2 Binary files /dev/null and b/figure/proportion-preferred_model-1.png differ diff --git a/figure/proportion_preferred_figure-1.png b/figure/proportion_preferred_figure-1.png new file mode 100644 index 0000000..6083101 Binary files /dev/null and b/figure/proportion_preferred_figure-1.png differ diff --git a/img/0279c16e978ead0d8251c5d7a0bba30847ed7d74.png b/img/0279c16e978ead0d8251c5d7a0bba30847ed7d74.png new file mode 100644 index 0000000..7d8a17c Binary files /dev/null and b/img/0279c16e978ead0d8251c5d7a0bba30847ed7d74.png differ diff --git a/img/0e1a80602affd399473a766872e33749a2f0c1f7.png b/img/0e1a80602affd399473a766872e33749a2f0c1f7.png new file mode 100644 index 0000000..7b48cf2 Binary files /dev/null and b/img/0e1a80602affd399473a766872e33749a2f0c1f7.png differ diff --git a/img/47331530ca00baee05e1597e872ae0a538c06ede.png b/img/47331530ca00baee05e1597e872ae0a538c06ede.png new file mode 100644 index 0000000..9f6010c Binary files /dev/null and b/img/47331530ca00baee05e1597e872ae0a538c06ede.png differ diff --git a/img/54eb0a21b143a53b6199a869d7a228ad7d158e57.png b/img/54eb0a21b143a53b6199a869d7a228ad7d158e57.png new file mode 100644 index 0000000..6083101 Binary files /dev/null and b/img/54eb0a21b143a53b6199a869d7a228ad7d158e57.png differ diff --git a/img/747344365b663c473e0d3ee0f87e35b9ab6ca8cb.png b/img/747344365b663c473e0d3ee0f87e35b9ab6ca8cb.png new file mode 100644 index 0000000..efddd98 Binary files /dev/null and b/img/747344365b663c473e0d3ee0f87e35b9ab6ca8cb.png differ diff --git a/rapport.pdf b/rapport.pdf index 7b4f9db..bf1389f 100644 Binary files a/rapport.pdf and b/rapport.pdf differ