Adds simple tweet search engine UI

.gitignore

@@ -162,3 +162,4 @@ cython_debug/
 
 
 data/tfidf_matrix.pckl
+data/tfidf_vectorizer.pckl

app.py

@@ -13,9 +13,9 @@ footer_html: str = load_html_str_from_file(os.path.join("www", "footer.html"))
 
 def nav_controls() -> List[NavSetArg]:
     return [
-        ui.nav(ui.h5("Intro"), mod_welcome.welcome_ui("Intro"), value="intro"),
+        ui.nav(ui.h5("Intro"), mod_welcome.welcome_ui("intro"), value="intro"),
         ui.nav(ui.h5("Analyse"), "Analyse"),
-        ui.nav(ui.h5("Suchmaschine"), mod_searchable.searchable_ui("Suchmaschine")),
+        ui.nav(ui.h5("Suchmaschine"), mod_searchable.searchable_ui("search_engine"), value="search_engine"),
         ui.nav_control(
             ui.a(
                 ui.h5("AG-Link"),
@@ -58,7 +58,8 @@ app_ui = ui.page_navbar(
 
 
 def server(input: Inputs, output: Outputs, session: Session):
-    mod_welcome.welcome_server("Intro")
+    mod_welcome.welcome_server("intro")
+    mod_searchable.searchable_server("search_engine")
 
 
 static_dir = Path(__file__).parent / "www"

src/mod_searchable.py

@@ -1,15 +1,93 @@
 from shiny import module, ui, render, Inputs, Outputs, Session
+from sklearn.feature_extraction.text import TfidfVectorizer
+from sklearn.metrics.pairwise import cosine_similarity
+import pandas as pd
+import numpy as np
+import pickle
+
+tfidf_matrix_path = "data/tfidf_matrix.pckl"
+tfidf_vectorizer_path = "data/tfidf_vectorizer.pckl"
+relevance_score_path = "data/tweet_relevance.json"
+tweets_path = "data/tweets_all_combined.csv"
+
+
+print("Loading data from storage")
+tweets = pd.read_csv(tweets_path)
+relevance_score = pd.read_csv(relevance_score_path)
+
+tfidf_matrix = None
+with open(tfidf_matrix_path, "rb") as f:
+    tfidf_matrix = pickle.load(f)
+
+tfidf_vectorizer: TfidfVectorizer = None
+with open(tfidf_vectorizer_path, "rb") as f:
+    tfidf_vectorizer = pickle.load(f)
+
+
+tweets["relevance_score"] = relevance_score["relevance_score"]
+tweets = tweets.drop(["user_id", "measured_at", "tweet_id"], axis=1)
+
+
+def search_query(query: str, limit: int = 5) -> pd.DataFrame:
+    query_vec = tfidf_vectorizer.transform([query])
+    similarity = cosine_similarity(query_vec, tfidf_matrix).flatten()
+
+    filtered = np.where(similarity != 0)[0]
+    indices = np.argsort(-similarity[filtered])
+    correct_indices = filtered[indices]
+    result = tweets.iloc[correct_indices]
+
+    if not len(result):
+        return None
+
+    overall = result['relevance_score'] * similarity[correct_indices]
+    return result.loc[overall.sort_values(ascending=False).index].head(limit)
 
 
 @module.ui
 def searchable_ui():
-    return ui.div("hello world")
+    return ui.div(
+        ui.h2("Tweet Suchmaschine"),
+        ui.input_text("search_input", "Suche:", placeholder="Gebe Suchterm ein"),
+        ui.HTML("<br>"),
+        ui.output_ui(id="searchable_tweet_ui"),
+    )
 
 
 @ module.server
-def searchable_server(input: Inputs, output: Outputs, session: Session, starting_value=0):
+def searchable_server(input: Inputs, output: Outputs, session: Session):
     @output
     @render.ui
-    def dataset_infos():
+    def searchable_tweet_ui():
 
-        return ui.markdown("")
+        query = input.search_input()
+
+        result_pd = search_query(query, 15)
+
+        style = "text-align: center; padding-top: 0.5em;"
+        tweet_ui = ui.page_fluid()
+
+        if result_pd is None:
+            return ui.div(
+                ui.h5("Keine Ergebnisse gefunden!")
+            )
+
+        # iterating over dataframe is bad but needed
+        for idx, row in result_pd.iterrows():
+            tweet_ui.append(
+                ui.div(
+                    ui.row(
+                        ui.column(9, ui.markdown(f"**{row['user_name']}** *@{row['handle']}*"), style=style),
+                        ui.column(3, ui.p(f"{row['created_at']}"), style=style),
+                    ),
+                    ui.row(
+                        ui.column(12, ui.HTML(str(row["tweet_text"]).replace("\\n", "<br>")), style=style + "font-size: 20px; padding:1em;"),
+                    ),
+                    ui.row(
+                        ui.column(3, ui.p(f"👍 {row['like_count']}"), style=style),
+                        ui.column(3, ui.p(f"⟲ {row['retweet_count']}"), style=style),
+                        ui.column(3, ui.p(f"↪ {row['reply_count']}"), style=style),
+                        ui.column(3, ui.p(f"💬 {row['quote_count']}"), style=style),
+                    ), style="border: 1px solid #954; margin-bottom: 1em;"))
+
+        return tweet_ui

src/mod_welcome.py

@@ -18,7 +18,6 @@ def welcome_ui():
 
             [0]: https://ag-link.xyz
             """),
-        # ui.output_text("dataset_infos"),
         ui.output_ui("dataset_infos"),
         ui.h3("Ursprung der Idee"),
         ui.markdown("""
@@ -52,7 +51,7 @@ with open("data/general_analysis_results.json", "r") as f:
 
 
 @ module.server
-def welcome_server(input, output, session, starting_value=0):
+def welcome_server(input, output, session):
     @output
     @render.ui
     def dataset_infos():