Kushal Shah-Machine Learning-Self assessment

Technical Area:

  • Scrum Workflow
  • Web scraping using BeautifulSoup
  • Tag Prediction using tf-idf
  • Used Libraries such as pandas, numpy, scikit-learn, ntlk

Tools:

  • Miro
  • Google Colab
  • Jupyter Notebook
  • GitHub
  • Python
  • Kaggle

Soft Skills:

  • Collaboration
  • Time Management
  • Perform background research

Achievement Highlights

  • Started implementing text processing and Exploratory data analysis on Kaggle data having 10000 rows of Stack-Overflow data.
  • Implemented Tf-idf on the same processed data.

Meetings attended

  • All of the Weekly meetings
  • Project Party on Wednesday

Goals for the Upcoming Week

  • To have the tag predictor model ready.
  • Prepare end to end working model by integrating work done by all the teams.

Tasks Done

  • Researched on Tag prediction by going through a multiple blogs and videos.
  • Create model for tag prediction on StackOverflow

Technical Area:

  • Scrum Workflow
  • Web scraping using BeautifulSoup
  • Tag Prediction using tf-idf
  • Used various classifiers such as logistic regression, naives bayes to build the model.
  • used roc curves, roc_auc_score, recall score, precision score.

Tools:

  • Miro
  • Google Colab
  • Jupyter Notebook
  • GitHub
  • Python
  • Kaggle

Soft Skills:

  • Collaboration
  • Time Management
  • Perform background research

Achievement Highlights

  • Started implementing text processing and Exploratory data analysis on Kaggle data having 10000 rows of Stack-Overflow data.
  • Implemented Tf-idf on the same processed data.
  • Used various classifiers such as logistic regression, naives bayes to build the model.
  • Generated roc curves, roc_auc_score, recall score, precision score to compare accuracy of the models.

Meetings attended

  • All of the Weekly meetings

Tasks Done

  • Researched on Tag prediction by going through a multiple blogs and videos.
  • Create model for tag prediction on data scraped by other team members.
  • Gathered information about the working of various machine learning models and how to compare their accuracies.
  • Researched about Bert implementation on the same scraped data to evaluate the difference between tf-idf and bert approach.