A_s12 - Machine Learning (Level 1) Pathway

Certificates Summer 2021 Certificates Machine Learning Pipeline (NLP) (CE) 2021
1

Technical Area

  • Learned to scrape data from websites

  • Learned to store data in a .csv file

  • Got a better understanding of how to use Python and BeautifulSoup

Tools

  • Python

  • BeautifulSoup

  • VSCode

Soft Skills

  • Better understand how websites are made

  • Better understand how to find specific information about a website from HTML tags

Achievements

  • Scraped data off of a DiscourseHub Community Forum

  • Cleaned HTML tags to find the title, tags, date, number of views, and number of replies on each post

  • Stored the data in a .csv file

Tasks Completed

  • Scraped data from DiscourseHub Forums

  • Stored data in a .csv file

Goals

  • Perform EDA to analyze the data
2

Technical Area

  • Fixed new issues with the code that I came across
  • Cleaned the data to make it easier to analyze
  • Used the Pandas library to analyze the data

Tools

  • Python
  • BeautifulSoup
  • VSCode
  • Pandas

Soft Skills

  • Understood how cleaning data made it easier to analyze
  • Learned how data is analyzed and visualized to find similar posts

Achievements

  • Improved the code to deal with emojis that weren’t being stored in a .csv file
  • Used Python libraries to analyze data
  • Used Python libraries to find posts that are similar to each other

Tasks Completed

  • Fixed code that stopped working due to a new post with an emoji being added
  • Analyzed and visualized data

Goals

  • Create a recommender system
  • Gather more data and overall improve the code
3

Technical Area

  • Fixed issues I had from previous module
  • Visualized the data to see how similar posts were
  • Learned ways to check for similarity between texts

Tools

  • Python
  • VSCode
  • Pandas
  • Pytorch
  • Matplotlib and Numpy

Soft Skills

  • Understood how to find similarity between words
  • Understood how recommender systems typically work

Achievements

  • Visualized the data to find similar posts
  • Created a system to give the 10 posts most similar to the one being seen

Tasks Completed

  • Fixed code that stopped working due to a new post with an emoji being added
  • Analyzed and visualized data

Goals

  • Continue working on the recommender system
  • Improve the methods of recommending posts
  • Gather more data to improve the recommender system
4

Technical Area

  • Improved the recommender system by providing more data
  • Fixed a few new issues I had found
  • Visualized the new data that I scraped

Tools

  • Python
  • VSCode
  • Pandas
  • Pytorch
  • Matplotlib and Numpy

Soft Skills

  • Further understood how recommender systems work
  • Learned how data is analyzed and visualized to find similar posts

Achievements

  • Found good recommender systems that reliably provide posts that are similar to the original
  • Used the better recommender systems to show me similar posts

Tasks Completed

  • Compared the results of different recommender systems
  • Found a few ways of getting good recommended posts

Goals

  • Further test the systems to find the one that works best
5

Technical Area

  • Worked with the rest of my group to discuss the different recommender systems we had
  • Tested the different recommender systems in order to determine which ones were better
  • Decided on a recommender system to use for the next module

Tools

  • Python
  • VSCode
  • Pandas
  • Pytorch
  • Matplotlib and Numpy

Soft Skills

  • Worked with my team to compare recommender systems
  • Tested out our recommender systems in order to understand what makes some better

Achievements

  • Found a recommender system that we felt gave the best results

Tasks Completed

  • Compared the results of different recommender systems along with the rest of my team
  • Started finalizing our project and putting everything together

Goals

  • Complete the last module by putting everything together to create a final project
6

Technical Area

  • Finalized anything with the recommender system that still needed to be fixed
  • Wrapped the model into an API
  • Built the final app

Tools

  • Docker
  • Streamlit

Soft Skills

  • Learned how to put together an app using what we have made

Achievements

  • Finished all work on the model
  • Finalized the project

Tasks Completed

  • Made a few improvements to our model from the last module
  • Created our finalized app containing the model