Detect spam messages with a text classifier

Research question

You write thisOne question you can answer with data, narrow enough to finish in the time you have.

Hypothesis

You write thisWhat you expect to find, and why.

Variables

You write thisWhat you change or compare, what you measure, and what you hold constant.

Methodology

Test how well the model works on messages from a different year or country and discuss why it degrades.

Data collection

You write thisSay what you will record, how many times, and how you will keep the records safe.

Analysis

You write thisChoose the table, chart or test that answers the question; report averages with their spread.

Limitations

You write thisList what could have affected the result: small sample, instrument limits, things you could not control.

Literature review

Find five to eight sources (your textbook, review articles, reports) and note what each says about your question. Group them by idea, not one after another, and end with what is still not known.

References

List every source you used in one style throughout (author, year, title, where it was published, link and the date you opened it).

Use this as the outline of your README or case study.

Problem

A public dataset of about 5,500 labelled SMS messages is enough to train a classifier that catches most spam.

Why it matters

It is a real product feature, and it teaches why accuracy alone is misleading when one class is rare.

What was built

A classifier with its metrics, the top spam words and examples of mistakes.

Technology

pandas, Python, scikit-learn

Architecture

You write thisOne diagram: the parts and how data moves between them.

Implementation

You write thisThe 6 build steps in your own words, with one code or design decision you are proud of.

Challenges

You write thisThe hardest bug or decision, and how you got past it.

Results

Type five messages of your own, including a tricky one, and check the predictions.

Demo and code

You write thisA link to a live demo or a short video, and to the GitHub repository.

Lessons learned

You write thisWhat you would do differently next time.

Future improvements

Compare naive Bayes with logistic regression and TF-IDF features.

Problem

A public dataset of about 5,500 labelled SMS messages is enough to train a classifier that catches most spam.

Target user

You write thisOne specific person: who they are and when they would use this.

Solution

Train a model on a public SMS dataset to separate spam from normal messages.

First version (MVP)

Count spam words by hand in 50 messages and write a rule-based filter.

Tech stack

pandas, Python, scikit-learn

Architecture

You write thisA quick sketch of the parts and how they connect.

Demo

Type five messages of your own, including a tricky one, and check the predictions.

Stretch goals

Compare naive Bayes with logistic regression and TF-IDF features.

Team roles

FrontendBackendData or modelDesignTestingDocumentationPresentation

Plan by length