app.py

import streamlit as st
import re

# MUST BE FIRST STREAMLIT COMMAND
st.set_page_config(
    page_title="Learn NLP from Scratch",
    page_icon="🧠",
    layout="wide"
)

# ---------------------------
# Helper Functions
# ---------------------------

def tokenize(text):
    return re.findall(r"\b\w+\b", text.lower())

STOPWORDS = {
    "is", "am", "are", "the", "a", "an", "and", "or", "in", "on", "at", "to", "of"
}

def remove_stopwords(tokens):
    return [t for t in tokens if t not in STOPWORDS]

def simple_stem(word):
    for suffix in ["ing", "ed", "s"]:
        if word.endswith(suffix):
            return word[:-len(suffix)]
    return word

def stem_tokens(tokens):
    return [simple_stem(t) for t in tokens]

def simple_pos_tag(tokens):
    tagged = []
    for word in tokens:
        if word.endswith("ing"):
            tagged.append((word, "VERB"))
        else:
            tagged.append((word, "WORD"))
    return tagged

# ---------------------------
# App UI
# ---------------------------

st.title("Natural Language Processing (NLP) – From Basics to Practice")

st.write(
    "This app explains the **NLP lifecycle**, **core techniques**, and provides a "
    "**hands-on playground** to understand how text is processed by machines."
)

tabs = st.tabs([
    "NLP Lifecycle",
    "NLP Techniques",
    "NLP Playground",
    "NLP Roadmap"
])

# ---------------------------
# NLP Lifecycle
# ---------------------------

with tabs[0]:
    st.header("NLP Lifecycle")

    st.markdown("""
    **1. Data Collection**  
    Collect text data such as reviews, emails, chats, or tweets.

    **Example:** Amazon product reviews.

    **2. Text Preprocessing**  
    Clean and prepare the text by removing noise.

    **Example:**  
    `I Love NLP!!!` → `i love nlp`

    **3. Feature Extraction**  
    Convert text into numerical form.

    **Example:** Bag of Words, TF-IDF.

    **4. Model Training**  
    Train a machine learning or deep learning model.

    **Example:** Spam detection model.

    **5. Evaluation**  
    Measure model performance.

    **Example:** Accuracy = 90%.

    **6. Deployment**  
    Use the model in real applications.

    **Example:** Chatbots, search engines.
    """)

# ---------------------------
# NLP Techniques
# ---------------------------

with tabs[1]:
    st.header("NLP Techniques")

    st.markdown("""
    **Tokenization**  
    Splits text into words.  
    *Example:* `I love NLP` → `['i', 'love', 'nlp']`  
    *Use:* Text preprocessing  
    *Advantage:* Easy to analyze text

    **Stopword Removal**  
    Removes common words like *is, the, and*.  
    *Use:* Reduces noise  
    *Advantage:* Improves performance

    **Stemming**  
    Converts words to root form.  
    *Example:* `playing → play`  
    *Use:* Search engines  
    *Advantage:* Smaller vocabulary

    **POS Tagging**  
    Identifies grammatical role of words.  
    *Example:* `learning → VERB`  
    *Use:* Grammar analysis  
    *Advantage:* Better sentence understanding
    """)

# ---------------------------
# NLP Playground
# ---------------------------

with tabs[2]:
    st.header("NLP Playground")

    text = st.text_area(
        "Enter text below",
        "I am learning Natural Language Processing"
    )

    if st.button("Run NLP"):
        tokens = tokenize(text)
        no_stop = remove_stopwords(tokens)
        stemmed = stem_tokens(no_stop)
        pos = simple_pos_tag(tokens)

        st.subheader("Tokens")
        st.write(tokens)

        st.subheader("After Stopword Removal")
        st.write(no_stop)

        st.subheader("After Stemming")
        st.write(stemmed)

        st.subheader("Simple POS Tagging")
        st.write(pos)

# ---------------------------
# NLP Roadmap
# ---------------------------

with tabs[3]:
    st.header("NLP Roadmap")

    st.markdown("""
    **Beginner**
    - Text cleaning
    - Tokenization
    - TF-IDF

    **Intermediate**
    - Machine learning models
    - Word embeddings
    - POS & NER

    **Advanced**
    - LSTM
    - Transformers
    - BERT, GPT

    **Applications**
    - Chatbots
    - Recommendation systems
    - Voice assistants
    """)