Spaces:
Sleeping
Sleeping
File size: 1,977 Bytes
887e4f5 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 |
import gradio as gr
import pandas as pd
from serpapi import GoogleSearch
# SERP API key (replace with your actual key)
SERP_API_KEY = "785988650046bf6eddbc597cbf87330e2d53f8a3bacb4bac62a90ab1ecfa2445"
def search_and_answer(question):
try:
# Step 1: Fetch search results from Google using SERP API
search_params = {
"q": question,
"hl": "en",
"gl": "us",
"api_key": SERP_API_KEY
}
search = GoogleSearch(search_params)
results = search.get_dict()
# Extract top 3 organic search results
extracted_results = []
for result in results.get("organic_results", [])[:3]:
extracted_results.append({
"title": result.get("title"),
"link": result.get("link"),
"snippet": result.get("snippet", "No description available.")
})
if not extracted_results:
return pd.DataFrame(columns=["Answer", "Source", "Confidence Score"])
# Step 2: Prepare final dataframe with sources and confidence scores
data = []
for i, res in enumerate(extracted_results):
confidence_score = round(1 - (i * 0.2), 2) # Simulated confidence score
data.append({
"Answer": res["snippet"],
"Source": res["link"],
"Confidence Score": confidence_score
})
df = pd.DataFrame(data)
return df
except Exception as e:
return pd.DataFrame({"Error": [str(e)]})
# Step 3: Create Gradio Interface
iface = gr.Interface(
fn=search_and_answer,
inputs=gr.Textbox(label="Ask a Question"),
outputs=gr.Dataframe(headers=["Answer", "Source", "Confidence Score"]),
title="AI-Powered Q&A System ",
description="Enter a question and get top 3 answers from web search with confidence scores."
)
# Launch the Gradio app with debug enabled
iface.launch(share=True, debug=True)
|