major update

app.py

@@ -1,9 +1,11 @@
 import streamlit as st
 from annotated_text import annotated_text
 from refined.inference.processor import Refined
+import nltk
+nltk.download('punkt')
 
 # Sidebar
-# st.sidebar.image("logo-wordlift.png")
+st.sidebar.image("logo-wordlift.png")
 
 # Initiate the model
 model_options = {"aida_model", "wikipedia_model_with_numbers"}
@@ -48,9 +50,23 @@ if text_input:
     annotations = []
     for wikidata_link, entity in entities_map.items():
         description = get_entity_description(wikidata_link, combined_entity_info_dictionary)
-        annotations.append((entity, wikidata_link, description))
-        st.write(entity + " , " + wikidata_link + " , " + description)
-        
+        annotations.append((entity, description, "#8ef"))
+
     # Annotate text with entities
     if submit_button:
-        annotated_text(*annotations)
+        # Split the input text into words
+        words = nltk.word_tokenize(text_input)
+
+        # Prepare a list to hold the final output
+        final_text = []
+
+        for word in words:
+            # If the word is an entity, annotate it
+            if word in entities_map.keys():
+                final_text.append((word, get_entity_description(word, combined_entity_info_dictionary), "#8ef"))
+            # If the word is not an entity, keep it as it is
+            else:
+                final_text.append(word)
+
+        # Pass the final_text to the annotated_text function
+        annotated_text(*final_text)