Skip to content
View saloni-jain-code's full-sized avatar
1 follower · 0 following

Achievements

Achievement: Pull Shark

Achievements

Achievement: Pull Shark

Highlights

  • Pro

Block or report saloni-jain-code

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
saloni-jain-code/README.md

Hi, I'm Saloni 👋

I'm a senior at Columbia University studying Computer Science and Applied Math. I'm passionate about solving real-world problems with machine learning and artificial intelligence. I'm also a strong self-learner, leader, and collaborator eager to apply my technical skills to build impactful solutions. Connect with me on Linkedin or contact me at sj3230@columbia.edu!

🎯 Featured Project: Topic Modeling for The Clorox Company

Check out my work on this topic modeling project for The Clorox Company! My team and I developed a several different topic modeling modules to analyze 670K+ consumer reviews, helping the company extract valuable insights to improve their products.

  • Tools: Python, Top2Vec, LSA, LDA2Vec, BERTopic, Llama3
  • What We Did: Tested Top2Vec, LSA, LDA2Vec, and BERTopic topic modeling methods. Preprocessed and vectorized text data, converted text to embeddings, did dimensionality reduction, clustered the text embeddings, and extracted top terns from each of the topic clusters. Top terms were passed through the Llama3 LLM to generate coherent topic labels. Evaluated topic labels thorugh similarity score between topic label and review and through qualitative evaluation.
  • Results: After testing all four methods, we found that LSA worked the best, and provided insights that helped guide data-driven improvements in product development. We developed modular, reusable code that allowed users to build, train, and evaluate a model within 2 lines of code, and also created a command line tool to make our code easy to use for future data engineers at Clorox.

🛠 Tech Stack:

  • Programming Languages: Python, JavaScript, Java, C
  • Libraries & Frameworks: React.js, Node.js, Express.js, scikit-learn, Pandas, NumPy, matplotlib, Langchain
  • Databases: MySQL, MongoDB

Pinned Loading

  1. AI-Support-Bot AI-Support-Bot Public

    Jupyter Notebook

  2. Article-Bias-Prediction Article-Bias-Prediction Public

    This is my final Break Through Tech project. The model predicts the political bias of a news article using TF-IDF vectorization and a Gradient Boosted Classifier.

    Jupyter Notebook

  3. Flashcardify_Final Flashcardify_Final Public

    JavaScript

  4. Inventory-Tracker Inventory-Tracker Public

    JavaScript

  5. Keeper Keeper Public

    Recreating Google Keep 📝

    JavaScript

  6. mixologist_robot mixologist_robot Public

    Python 1