Applied Machine Learning Portfolio¶

Megan Chastain

2026-04

This page summarizes my work on NLP projects.

NLP Techniques Implemented¶

Describe the NLP techniques you used. Examples:

Tokenization (word, sentence) Frequency analysis (unigram / n-gram) Text cleaning and normalization API-based text analysis (and JSON) Web scraping / content extraction from HTML Sentiment analysis (e.g., spaCy + SpacyTextBlob)

Systems and Data Sources¶

Describe what you analyzed:

Web pages, APIs, datasets, or documents Differences in structure (HTML, JSON, plain text) How you handled or cleaned messy or variable data

Pipeline Structure (EVTL)¶

Describe your pipeline using EVTL:

Extract (from source): how data was collected Validate: structure/content checks performed Transform: NLP processing steps Load (to sink): outputs (files, summaries, visualizations)

Signals and Analysis Methods¶

Describe what you computed and how:

word frequency context or co-occurrence keyword extraction special signals sentiment or subjectivity

Insights¶

Describe what your analysis revealed:

patterns, trends, or notable findings anything surprising or unexpected what made the results useful or meaningful

Representative Work¶

Provide links to 2–3 of your strongest projects.

For each project:

include a clickable link provide 1–2 sentences describing what it does and why it is representative

Skills¶

Be explicit about what you can do. Avoid general statements.

Examples:

Python data processing working with text data handling messy or inconsistent inputs structuring repeatable pipelines communicating results professionally with Markdown and visuals