push

app.py

@@ -241,7 +241,7 @@ examples = [
     ["What companies report earnings this week"],
     ["write me a health check on adobe"],
     ["Analyze the technical indicators for Apple"],
-    ["Build an intrinsic value model for Apple"],
+    ["give me the current sentiment score for Apple"],
     ["Make a table of Apple's stock price for the last 3 days"],
     ["What is Apple's PE ratio and how does it compare to other companies in consumer electronics"],
     ["How did Salesforce perform in its last earnings?"],

healthcheck.txt

@@ -169,4 +169,23 @@ A Recommendation Statement, if necessary, for areas needing attention.
 
 
 SMART SHEET OBJECTIVE:
-ANALYSE THE NEWS BROUGHT IN ON THE TOPIC IN IMMENSE DETAIL, ANALYSING EVERYTHING AND SIMULATING POSSIBLE OUTCOMES(WHERE NECCESSARY), EXPLAIN THE IMPLICATIONS OF THE NEWS AND WHAT MAY HAPPEN BECAUSE OF IT, ANALYSE IT IN DEPTH.
+ANALYSE THE NEWS BROUGHT IN ON THE TOPIC IN IMMENSE DETAIL, ANALYSING EVERYTHING AND SIMULATING POSSIBLE OUTCOMES(WHERE NECCESSARY), EXPLAIN THE IMPLICATIONS OF THE NEWS AND WHAT MAY HAPPEN BECAUSE OF IT, ANALYSE IT IN DEPTH.
+
+SENTIMENT SCORE OBJECTIVE:
+Analyze the retrieved data to produce a Quantineuron sentiment score for [Company Name] on a scale from 0 to 100, where 100 indicates very positive sentiment and 0 indicates very negative sentiment.
+
+1. **Identify Segments**: Divide the retrieved text into sections for news, social media, and analyst opinions based on context cues such as "news," "analyst," "social media," "Twitter," "buy," "sell," or "forecast."
+2. **Sentiment Scoring**:
+    - **News**: Analyze sentiment in news sections for positivity, negativity, or neutrality regarding the company’s stock. Assign a sentiment score for news between 0-100.
+    - **Social Media**: Gauge sentiment in social media mentions (if present) by analyzing terms like "bullish," "bearish," or other sentiment expressions. Score social sentiment between 0-100.
+    - **Analyst Opinions**: Identify sections with analyst opinions or stock ratings (e.g., "buy," "hold," "sell") and score them between 0-100 based on sentiment indicators.
+3. **Score Aggregation**:
+    - Aggregate the sentiment scores for each segment, assigning higher weights to news and analyst sections, and calculate an overall Quantineuron sentiment score between 0-100.
+4. **User-Friendly Output**:
+    - Display the Quantineuron sentiment score alongside a brief summary of key sentiment indicators (e.g., "positive analyst ratings," "mixed social media reactions").
+    - Optionally, provide a suggestion based on the sentiment score, e.g., "Consider this score for understanding potential market sentiment in the short term."
+
+**Output Format**:
+- "Quantineuron sentiment score for [Company Name]: [Score] out of 100."
+- "Summary: [Brief sentiment highlights]."
+- Optional Suggestion: "[Guidance based on sentiment]."

search_requirements.txt

@@ -5,4 +5,7 @@ WHEN THE USER ASKS FOR A HEALTH CHECK:
 search for the latest metrics on the company
 
 WHEN THE USER ASKS FOR A SMART SHEET:
-search for the latest news on the event
+search for the latest news on the event
+
+WHEN THE USER ASKS FOR A SENTIMENT SCORE:
+SEARCH FOR THE SENTIMENT DATA ON THE COMPANY WHETHER THAT BE NEWS SENTIMENT, SOCIAL MEDIA SENTIMENT OR MARKET SENTIMENT TOWARDS THE STOCK