add descriptions to language typology and cross-domain pages

pages/3_Language_Typology.py

@@ -14,6 +14,11 @@ def get_typology_df():
 st.set_page_config(page_title="Language Typology", page_icon="📈")
 
 st.markdown("# Language Typology")
+st.write("""\
+Languages can be described using hundreds of linguistic features. [World Atlas of Language Structures](https://doi.org/10.5281/zenodo.7385533) lists almost 200 different features. Since our work focuses on sentiment classification, we select 10 features that seem to be the most relevant to the task of sentiment expression.  
+
+The table below presents the languages included in the MMS corpus, their family, genus and their values for the selected linguistic features.  
+You can use the **'Add filters'** checkbox to filter the table by any of the columns.""")
 
 df = get_typology_df()
 

pages/{4_Domain_Transfer.py → 4_Cross-domain_Transfer.py}

@@ -17,6 +17,10 @@ st.markdown(f"# {TITLE}")
 
 df = get_domain_transfer_data()
 
+st.write("""\
+    The figure below shows the results of an experiment comparing the effectiveness of transfer learning between data modalities. As expected, when models are tested against datasets from the same domain (news, social media, reviews), the average performance is much higher than on out-of-domain datasets. What is striking is the visible deterioration of the performance of models trained in the news domain. When knowledge transfer happens between social media and review domains, the average  performance of models stays relatively the same."""
+)
+
 st.plotly_chart(
     px.box(
         df,