ELtrebolt - Machine Learning Pathway

Certificates July 2020 Certificates Machine Learning Pipeline (NLP)Self-Assessments
1

Weeks 1-2: July 20th – July 31st

Things Learned

Technical:

  • Python (Basic) – For Loops, Lists, Conditionals
  • Data Mining (Basic) – Beautiful Soup and Selenium
  • Pandas Dataframe

Tools:

  • Code Management: Git, Github
  • Project Management: Slack, Asana
  • Code Editing: Anaconda, Python, Visual Studio Code

Soft Skills:

  • Learning Mindset
  • Online Collaboration
  • Problem Solving & Troubleshooting

Achievement Highlights

  • Worked with a team in scraping and combining data from a website
  • Used Beautiful Soup and Selenium for web scraping

Meetings Attended

  • 2-3 Weekly ML Team 6 meetings
  • 2 Weekly Codecademy Group meetings

Goals for Upcoming Meetings

  • Make observations and begin data pre-processing
  • Learn basic machine learning models for recommender systems

Tasks Done

  • Plan Data Collection: Collaborated with a small group to record observations about the Codecademy forum. Made a report including the metrics, navigation, display, topic elements, and useful categories of the forum.

  • Data Collection: Scraped a portion of the categories from the Codecademy forum and put together each topic’s title, category, tags, content, and comments into a csv file. Collaborated with the team in scraping the right elements and optimizing code.

2

Weeks 3-4: August 3rd – August 14th

Things Learned

Technical:

  • Python (Basic) – Dictionaries, Functions
  • Data Pre-Processing (Basic) – Tokenize, Rake, Lemmatizer
  • Machine Learning Models – TF-IDF score, Vectors, Cosine Similarity

Tools:

  • Jupyter Notebook
  • Microsoft Excel

Soft Skills:

  • Learning Mindset
  • Problem Solving & Troubleshooting
  • Resourcefulness

Achievement Highlights

  • Performed data cleaning
  • Began working with Machine Learning models

Meetings Attended

  • 2 Weekly ML Team 6 meetings

Goals for Upcoming Meetings

  • Collaborate with team to create a successful machine learning recommendation model

Tasks Done

  • Data Pre-Processing: Utilized tokenization, stopword removal, stemming / lemmatization, and removal of unwanted characters to clean data. Formatted data into a “bag of words” style format.

  • Learned Machine Learning Models: Researched, took notes on, and began experimenting with TF-IDF, Simple Transformers, and BERT models for a recommendation system.

3

Weeks 5-6: August 17th – August 28th

Things Learned

Technical:

  • Machine Learning Models – BERT, Sentence Transformers

Tools:

  • Google Colab

Soft Skills:

  • Public Speaking / Presenting
  • Professionalism

Achievement Highlights

  • Implemented recommender system
  • Completed final presentation

Meetings Attended

  • 2 Weekly ML Team 6 meetings

Tasks Done

  • Implemented Recommender System: Created a machine learning recommendation model using Sentence Bert to perform Semantic Search, returning topics similar to any given query.

  • Completed Final Presentation: Worked on, practiced, and presented an overview of the project for the team’s leads and STEM-Away’s mentors. Included phases such as Team Setup, Data Collection, Data Pre-Processing, and Data Modeling.