Spaces:
Running
Running
RamiIbrahim
commited on
Commit
•
bac5e0b
1
Parent(s):
af08641
Stable version 1
Browse files
app.py
CHANGED
@@ -7,28 +7,60 @@ model = joblib.load('tunisian_arabiz_sentiment_analysis_model.pkl')
|
|
7 |
vectorizer = joblib.load('tfidf_vectorizer.pkl')
|
8 |
|
9 |
def predict_sentiment(text):
|
10 |
-
# Transform the input text using the loaded vectorizer
|
11 |
text_vectorized = vectorizer.transform([text])
|
12 |
-
|
13 |
-
# Make prediction
|
14 |
prediction = model.predict(text_vectorized)[0]
|
15 |
-
|
16 |
-
# Convert prediction to sentiment
|
17 |
-
sentiment = "Positive" if prediction == 1 else "Negative"
|
18 |
-
|
19 |
-
# Get prediction probability
|
20 |
probabilities = model.predict_proba(text_vectorized)[0]
|
21 |
confidence = max(probabilities)
|
22 |
|
23 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
24 |
|
25 |
# Create Gradio interface
|
26 |
iface = gr.Interface(
|
27 |
fn=predict_sentiment,
|
28 |
inputs=gr.Textbox(lines=3, placeholder="Enter Tunisian Arabiz text here..."),
|
29 |
-
outputs=
|
|
|
|
|
|
|
|
|
|
|
30 |
title="Tunisian Arabiz Sentiment Analysis",
|
31 |
-
description="
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
32 |
)
|
33 |
|
34 |
# Launch the interface
|
|
|
7 |
vectorizer = joblib.load('tfidf_vectorizer.pkl')
|
8 |
|
9 |
def predict_sentiment(text):
|
|
|
10 |
text_vectorized = vectorizer.transform([text])
|
|
|
|
|
11 |
prediction = model.predict(text_vectorized)[0]
|
|
|
|
|
|
|
|
|
|
|
12 |
probabilities = model.predict_proba(text_vectorized)[0]
|
13 |
confidence = max(probabilities)
|
14 |
|
15 |
+
sentiment = "Positive" if prediction == 1 else "Negative"
|
16 |
+
|
17 |
+
return {
|
18 |
+
"Sentiment": sentiment,
|
19 |
+
"Confidence": f"{confidence:.2f}",
|
20 |
+
"Explanation": f"The model predicts this text is {sentiment.lower()} with {confidence:.2%} confidence."
|
21 |
+
}
|
22 |
+
|
23 |
+
# Example texts
|
24 |
+
examples = [
|
25 |
+
["3ajbetni barcha el film hedhi"],
|
26 |
+
["ma7abitch el akla mte3 el restaurant"],
|
27 |
+
["el jaw fi tounes a7la 7aja"],
|
28 |
+
["ennes el kol za3nin w ma3andhomch flous"]
|
29 |
+
]
|
30 |
|
31 |
# Create Gradio interface
|
32 |
iface = gr.Interface(
|
33 |
fn=predict_sentiment,
|
34 |
inputs=gr.Textbox(lines=3, placeholder="Enter Tunisian Arabiz text here..."),
|
35 |
+
outputs={
|
36 |
+
"Sentiment": gr.Label(label="Predicted Sentiment"),
|
37 |
+
"Confidence": gr.Label(label="Confidence Score"),
|
38 |
+
"Explanation": gr.Textbox(label="Explanation")
|
39 |
+
},
|
40 |
+
examples=examples,
|
41 |
title="Tunisian Arabiz Sentiment Analysis",
|
42 |
+
description="""
|
43 |
+
This model predicts the sentiment of Tunisian Arabiz text as either Positive or Negative.
|
44 |
+
|
45 |
+
Tunisian Arabiz is a form of writing Arabic (specifically Tunisian dialect) using Latin characters and numbers.
|
46 |
+
|
47 |
+
Example:
|
48 |
+
- "3ajbetni" means "I liked it"
|
49 |
+
- "7aja" means "thing"
|
50 |
+
|
51 |
+
Try the examples below or enter your own text!
|
52 |
+
""",
|
53 |
+
article="""
|
54 |
+
<div style="text-align: center;">
|
55 |
+
<img src="https://upload.wikimedia.org/wikipedia/commons/c/ce/Flag_of_Tunisia.svg" alt="Tunisian Flag" style="width:150px;"/>
|
56 |
+
</div>
|
57 |
+
<h3>About the Model</h3>
|
58 |
+
<p>This sentiment analysis model was trained on a dataset combining TuniziDataset and the Tunisian Dialect Corpus.
|
59 |
+
It uses TF-IDF vectorization for feature extraction and Logistic Regression for classification.</p>
|
60 |
+
<h3>Limitations</h3>
|
61 |
+
<p>The model may not perform well on very colloquial expressions or new slang terms not present in the training data.
|
62 |
+
It's also important to note that sentiment can be nuanced and context-dependent, which may not always be captured by this model.</p>
|
63 |
+
"""
|
64 |
)
|
65 |
|
66 |
# Launch the interface
|