Check institution frequencies

WOS/wos_analysis/u_institution_frequency_exploration.py

@@ -0,0 +1,14 @@
+#%% Compare the frequency of the inst. names with the wos-analysis on the web interface
+import pandas as pd
+
+# %% 
+wos_inst_df = pd.read_excel("../wos_processed_data/wos_institution_locations_harmonized.xlsx")
+
+# %% Get rid of the duplicate institution name entries in each individual publication
+wos_inst_df["id_inst"] = wos_inst_df['Institution'] + ";" + wos_inst_df["UT (Unique WOS ID)"]
+wos_pubunique_inst = wos_inst_df["id_inst"].unique()
+wos_pubunique_inst = [x.split(";")[0] for x in wos_pubunique_inst]
+
+# %% Calc. frequencies, and get an output
+wos_inst_freqdf = pd.DataFrame(wos_pubunique_inst).value_counts().rename_axis('institution').reset_index(name='frequency')
+wos_inst_freqdf.to_csv("../wos_processed_data/wos_pubunique-institution_frequency.csv", index=False)