Sentiment analysis with scikit-learn in ML Python - Model Pipeline Trace

Model Pipeline - Sentiment analysis with scikit-learn

This pipeline takes text reviews and teaches a computer to tell if the feeling is positive or negative. It cleans the text, turns words into numbers, trains a simple model, and then checks how well it learned.

Data Flow - 5 Stages

1Raw text data

1000 rows x 1 column→Load text reviews with sentiment labels→1000 rows x 2 columns

['I love this product!', 'This is terrible.'], [1, 0]

↓

2Text cleaning and vectorization

1000 rows x 1 column→Convert text to numbers using CountVectorizer→1000 rows x 5000 columns

[0, 1, 0, ..., 2, 0, 1]

↓

3Train/test split

1000 rows x 5000 columns→Split data into 800 training and 200 testing rows→800 rows x 5000 columns (train), 200 rows x 5000 columns (test)

Train features shape: (800, 5000), Test features shape: (200, 5000)

↓

4Model training

800 rows x 5000 columns→Train Logistic Regression model on training data→Trained model

Model learns weights for each word feature

↓

5Model evaluation

200 rows x 5000 columns→Predict sentiment on test data and calculate accuracy→Accuracy score (scalar)

Accuracy = 0.85

Training Trace - Epoch by Epoch

Loss
0.5 |****
0.4 |****
0.3 |******
0.2 |*******
     1  2  3 Epochs

Epoch	Loss ↓	Accuracy ↑	Observation
1	0.45	0.75	Model starts learning, accuracy is moderate
2	0.3	0.82	Loss decreases, accuracy improves
3	0.25	0.85	Model converges with good accuracy

Prediction Trace - 3 Layers

Layer 1: Text vectorization

Layer 2: Logistic Regression prediction

Layer 3: Threshold decision

Model Quiz - 3 Questions

What does the vectorizer do in this pipeline?

A Splits data into training and testing sets

B Calculates the accuracy of the model

C Turns text into numbers representing word counts

D Predicts sentiment from the text

Key Insight

This visualization shows how text data is turned into numbers so a simple model can learn to tell positive from negative reviews. Watching loss go down and accuracy go up helps us know the model is learning well.