Update app.py

app.py

@@ -11,8 +11,6 @@ from pytz import timezone
 import plotly.graph_objects as go
 from sklearn.linear_model import LinearRegression
 from vaderSentiment.vaderSentiment import SentimentIntensityAnalyzer
-from textblob.sentimiento import PatternAnalyzer
-from spacy.lang.en import English
 from sklearn.model_selection import train_test_split
 from sklearn.ensemble import RandomForestClassifier
 from sklearn.metrics import accuracy_score
@@ -92,37 +90,27 @@ def get_crypto_news(api_key, crypto_symbol, articles_count=10):
     else:
         return []
 
-def analyze_news_sentiment(news, customized_thresholds=None, domain_lexicon=None):
-    if domain_lexicon is None:
-        # Use a default sentiment analyzer
-        analyzer = SentimentIntensityAnalyzer()
-    else:
-        # Use a custom sentiment analyzer with domain-specific lexicon
-        nlp = English()
-        nlp.add_pipe("textblob_sentiment", config={"lexicon": domain_lexicon})
-        analyzer = nlp.get_pipe("textblob_sentiment")
-
-    if customized_thresholds is None:
-        customized_thresholds = {
-            "positive": 0.5,
-            "negative": -0.5,
-            "neutral": 0.1
-        }
-
+def custom_sentiment_analysis(news, domain_lexicon):
+    analyzer = SentimentIntensityAnalyzer()
     for article in news:
         title = article['title']
         description = article['description']
         sentiment_score = analyzer.polarity_scores(title + " " + description)
 
-        if sentiment_score['compound'] >= customized_thresholds['positive']:
+        # Use the domain-specific lexicon to adjust the sentiment score
+        for term, weight in domain_lexicon.items():
+            if term.lower() in (title + " " + description).lower():
+                sentiment_score['compound'] += weight
+
+        if sentiment_score['compound'] >= 0.5:
             article['sentiment'] = 'positive'
-        elif sentiment_score['compound'] <= customized_thresholds['negative']:
+        elif sentiment_score['compound'] <= -0.5:
             article['sentiment'] = 'negative'
         else:
             article['sentiment'] = 'neutral'
 
     return news
-
+    
 def train_price_prediction_model(data):
     X = data[['Open', 'High', 'Low', 'Volume']]
     y = data['Close']