Streamlit Spam Classification

A quite simple application that demonstrates the use of Naive Bayes algorithm for multinomial models, pandas for reading datasets and other sickitlearn models.

Requirements

• Python 3 +
• Pip (To Install packages)
• Streamlit
• Pandas
• Numpy
• Sickitlearn

Installation

Step 1

Clone this repository.

git clone https://github.com/grayoj/spam-detection.git

Install the Python Programming Language

Step 2 Using Pip, Install Streamlit which is the server for the application.

$ pip install streamlit

Step 3 Install pandas to read data sets

$ pip install pandas

Step 4 Install Sickitlearn modules that contains all necessary modules used in the project

$ pip install sklearn

Step 5 Install Numpy for mathematical functions.

$ pip install numpy

Step 6 Install the Natural language model toolkit.

$ pip install nltk

If you use Pylance, it should validate all imports made.

# Import String
import string
from sklearn.model_selection import train_test_split

# Import Streamlit.
import streamlit as st


# Import pandas to read CSV files.
import pandas as pd

# Import natural language toolkit.
import nltk

nltk.download('stopwords')
from nltk.corpus import stopwords

# Import Sickitlearn
import sklearn

# Import Naive Bayes Module
from sklearn.naive_bayes import MultinomialNB
from sklearn.feature_extraction.text import CountVectorizer
# Import module to display accuracy
from sklearn.metrics import accuracy_score, confusion_matrix, classification_report

The imports as above.

Run Application

cd spam-detection
streamlit run app.py

Done. Should be running on the localhost. MIT License.