Update app.py

app.py

@@ -1,27 +1,34 @@
 import streamlit as st
 from transformers import pipeline
 
+# App title
+st.set_page_config(page_title="Fake Job Detector", page_icon="🕵️‍♂️")
 st.title("Fake Job / Lie Detector")
+st.markdown("Enter a job description and check if it seems suspicious!")
 
-st.write(
-    "This app classifies a job description as potentially fake or real. "
-    "Runs entirely locally on CPU, no API key needed."
-)
+# Load small NLI model for zero-shot classification (CPU-friendly)
+@st.cache_resource
+def load_model():
+    return pipeline("zero-shot-classification", model="facebook/bart-large-mnli")
 
-# Use a small, CPU-friendly model
-classifier = pipeline("zero-shot-classification", model="facebook/bart-large-mnli", device=-1)
+classifier = load_model()
 
+# Input from user
 job_description = st.text_area("Enter the job description:")
 
+# Label options
+labels = ["Legitimate", "Suspicious", "Fake", "Scam"]
+
+# Button to check
 if st.button("Check Job"):
     if not job_description.strip():
-        st.warning("Please enter a job description.")
+        st.warning("Please enter a job description!")
     else:
-        # Define the candidate labels
-        labels = ["real", "fake"]
-        result = classifier(job_description, candidate_labels=labels)
-        top_label = result["labels"][0]
-        score = result["scores"][0]
-
-        st.write(f"Prediction: **{top_label.upper()}**")
-        st.write(f"Confidence: **{score*100:.2f}%**")
+        # Run zero-shot classification
+        results = classifier(job_description, candidate_labels=labels)
+        st.subheader("Prediction:")
+        # Show top label and scores
+        top_label = results["labels"][0]
+        score = results["scores"][0]
+        st.success(f"Most likely: **{top_label}** ({score*100:.2f}%)")
+        st.write("Full results:", results)