diff --git a/Retrieval/commons.py b/Retrieval/commons.py
index d9077e4..79bf029 100644
--- a/Retrieval/commons.py
+++ b/Retrieval/commons.py
@@ -63,7 +63,7 @@ class TextRankings:
         O = self.obj
         docs_ids = [doc_id for doc_id, query_id in O['qid'].items() if query_id == sample_id]
         texts = [O['text'][doc_id] for doc_id in docs_ids]
-        labels = [O['continent'][doc_id] for doc_id in docs_ids]
+        labels = [O[self.class_name][doc_id] for doc_id in docs_ids]
         if max_lines > 0 and len(texts) > max_lines:
             ranks = [int(O['rank'][doc_id]) for doc_id in docs_ids]
             sel = np.argsort(ranks)[:max_lines]
diff --git a/Retrieval/fifth.py b/Retrieval/fifth.py
index 790ffca..af06b2b 100644
--- a/Retrieval/fifth.py
+++ b/Retrieval/fifth.py
@@ -104,26 +104,33 @@ RANK_AT_K = -1
 REDUCE_TR = 50000
 qp.environ['SAMPLE_SIZE'] = RANK_AT_K
 
-data_path = './newExperimentalSetup'
-train_path = join(data_path, 'train3000samples.json')
+data_path = {
+    'first_letter_category': './first_letter_categoryDataset',
+    'continent': './newExperimentalSetup'
+}
+
+def scape_latex(string):
+    return string.replace('_', '\_')
 
 
 Ks = [10, 50, 100, 250, 500, 1000, 2000]
 # Ks = [500]
 
-for CLASS_NAME in ['continent']: #, 'gender', 'gender_category', 'occupations', 'source_countries', 'source_subcont_regions', 'years_category', 'relative_pageviews_category']:
+for CLASS_NAME in ['first_letter_category']: #['continent']: #, 'gender', 'gender_category', 'occupations', 'source_countries', 'source_subcont_regions', 'years_category', 'relative_pageviews_category']:
+
+    train_path = join(data_path[CLASS_NAME], 'train3000samples.json')
 
     tfidf, classifier_trained = qp.util.pickled_resource(f'classifier_{CLASS_NAME}.pkl', train_classifier)
     trained=True
 
-    experiment_prot = RetrievedSamples(data_path,
+    experiment_prot = RetrievedSamples(data_path[CLASS_NAME],
                                load_fn=load_json_sample,
                                vectorizer=tfidf,
                                max_train_lines=None,
                                max_test_lines=RANK_AT_K, classes=classifier_trained.classes_, class_name=CLASS_NAME)
 
     method_names = [name for name, *other in methods()]
-    benchmarks = [f'{CLASS_NAME}@{k}' for k in Ks]
+    benchmarks = [f'{scape_latex(CLASS_NAME)}@{k}' for k in Ks]
     table_mae = Table(benchmarks, method_names, color_mode='global')
     table_mrae = Table(benchmarks, method_names, color_mode='global')
 
@@ -158,8 +165,9 @@ for CLASS_NAME in ['continent']: #, 'gender', 'gender_category', 'occupations',
             pbar.set_description(f'{method_name}')
 
         for k in Ks:
-            table_mae.add(benchmark=f'{CLASS_NAME}@{k}', method=method_name, values=mae_errors[k])
-            table_mrae.add(benchmark=f'{CLASS_NAME}@{k}', method=method_name, values=mrae_errors[k])
+
+            table_mae.add(benchmark=f'{scape_latex(CLASS_NAME)}@{k}', method=method_name, values=mae_errors[k])
+            table_mrae.add(benchmark=f'{scape_latex(CLASS_NAME)}@{k}', method=method_name, values=mrae_errors[k])
 
     table_mae.latexPDF('./latex', 'table_mae.tex')
     table_mrae.latexPDF('./latex', 'table_mrae.tex')