clustering algorithm, empty oid:id fixed

code/functions.py

@@ -388,6 +388,7 @@ def ocel_get_object_distances_sy(ocel, object_type, weights_per_attribute) -> di
     cflow_list = find_cflow(relations_df, activity_dict)
 
     # Sorts data frame and appends corresponding control flow
+    oid_filtered = oid_filtered.loc[oid_filtered['ocel:oid'] != '']
     oid_filtered.sort_values(by=['ocel:oid'], ascending=[True], inplace=True)
     oid_filtered = oid_filtered.assign(cflow=cflow_list)
 

code/main.py

@@ -10,16 +10,19 @@ import functions as f
 import cluster
 from sklearn import cluster as skcluster
 from sklearn_extra import cluster as cluster_extra
+from clusteval import clusteval
+from scipy.spatial import distance
 
 #
 # PARAMETERS
 p_ocel_file         = './data/ocel.jsonocel'
 p_mode              = 'existence' # all | existence
-p_object_type       = 'items' # object-type-name, options depend on the data
+p_object_type       = 'packages' # object-type-name, options depend on the data
 p_attr_weights = {
 } # attributes that are not given in the data are not used
+p_clusteval_mode = 'silhouette'
 p_clustering_mode   = 'kmeans' # optional, default: kmeans
-p_max_cluster_count = 10 # np.NaN # cluster-count (optional, default: np.NaN which leads to automatic k determination)
+# p_max_cluster_count = 25 # np.NaN # cluster-count (optional, default: np.NaN which leads to automatic k determination)
 p_ocel_file_type    = 'json' # json|xml
 p_graph_file_type   = 'svg' # svg|png
 # END PARAMETERS
@@ -33,8 +36,9 @@ print('p_mode:              "' + str(p_mode) + '".')
 print('p_object_type:       "' + str(p_object_type) + '".')
 print('p_attr_weights:')
 print(p_attr_weights)
+print('p_clusteval_mode:    "' + str(p_clusteval_mode) + '".')
 print('p_clustering_mode:   "' + str(p_clustering_mode) + '".')
-print('p_cluster_count:     "' + str(p_max_cluster_count) + '".')
+# print('p_max_cluster_count:     "' + str(p_max_cluster_count) + '".')
 print('p_ocel_file_type:    "' + str(p_ocel_file_type) + '".')
 print('p_graph_file_type:   "' + str(p_graph_file_type) + '".')
 print('-------------------------------------------------------')
@@ -78,28 +82,30 @@ index_to_oid_map = res['index']
 
 algo = algorithms[p_clustering_mode]
 
-try:
-    max_cluster_count = int(p_max_cluster_count)
-except:
-    max_cluster_count = 10
-assert max_cluster_count >= 2, 'cluster_count needs to be at least 2'
-assert max_cluster_count < len(index_to_oid_map), 'cluster_count needs to be less than the count of distinct objects in the ocel-data.'
+# try:
+#     max_cluster_count = int(p_max_cluster_count)
+# except:
+#     max_cluster_count = 25
+# assert max_cluster_count >= 2, 'cluster_count needs to be at least 2'
+# assert max_cluster_count < len(index_to_oid_map), 'cluster_count needs to be less than the count of distinct objects in the ocel-data.'
 
-cluster_count = cluster.determine_optimal_k(distance_matrix, algorithm=algo, k_max=max_cluster_count)
+# cluster_count = cluster.determine_optimal_k(distance_matrix, algorithm=algo, k_max=max_cluster_count)
+
+results = clusteval(evaluate = p_clusteval_mode).fit(distance_matrix)
+cluster_count = results['score']['clusters'].iloc[np.where(results['score']['score'] == results['score']['score'].max())[0][0]]
 
 algo.set_params(n_clusters=cluster_count)
 cluster_res = cluster.cluster_matrix(distance_matrix, algorithm=algo)
 
 df_clusters = pd.DataFrame({'cluster': cluster_res[1]}, index=index_to_oid_map) # creating dataframe
 df_clusters.index.name='ocel:oid' # setting index name for joining
-
 duration = datetime.datetime.now() - start_ts
 print('-------------------------------------------------------')
 print('                   OBJECT-CLUSTERING                   ')
 print('clustering-technique:')
 print(p_clustering_mode)
-print('max cluster count given:')
-print(max_cluster_count)
+# print('max cluster count given:')
+# print(max_cluster_count)
 print('cluster-count used:')
 print(cluster_count)
 print('object-cluster-dataframe:')