You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We deployed a recommendation system employing content-based filtering with real-world data. Initially, we conducted web scraping of UMass Faculty data using regular expressions for data preparation. Subsequently, we filtered the professors from the collected faculty data. To identify the research papers of these professors and determine their areas of interest, we leveraged Scholarly, ensuring accurate extraction of research papers. Next, we preprocessed and cleaned the data, generating word embeddings. Utilizing cosine similarity, we matched the word embeddings to the user input, thus creating an intuitive web application for the recommendation system.
Inspiration behind the Project
I really love doing research, but finding the right professor to work with can be tough. I tried searching online, but it took forever, and the information wasn't always accurate. Eventually, I had to rely on a connection from one of my current professors to find the right match. This got me thinking – what if there was a better way? What if there was a way for people interested in research to easily find professors who share their interests? This could help us connect faster and even boost interest in research, which could improve literacy rates.
🎉 Thank you for submitting your project to the BostonBridge Hackathon 2024! 🎉
Hi @shrutivarade!! We've successfully received your submission. Here's a quick rundown of what happens next:
Submission Review: Our judges will review your submission to ensure it meets all the hackathon criteria.
Presentation: You will be invited to present your project to the judges and other participants, you will have 3 minutes to present your project and 2 minutes for Q&A.
Notification: You'll be updated about the status of your submission as soon as the review is complete.
🔍 In the Meantime:
If you need to update your submission or have any queries, feel free to reply to this issue.
Keep an eye on this thread for any updates or further instructions.
🍀 Good Luck!
We're thrilled to see your innovative solution and wish you the best in the hackathon.
Name
Shruti Varade
Email Address
[email protected]
Email Addresses of Team Members
[email protected], [email protected], [email protected]
Project Description
We deployed a recommendation system employing content-based filtering with real-world data. Initially, we conducted web scraping of UMass Faculty data using regular expressions for data preparation. Subsequently, we filtered the professors from the collected faculty data. To identify the research papers of these professors and determine their areas of interest, we leveraged Scholarly, ensuring accurate extraction of research papers. Next, we preprocessed and cleaned the data, generating word embeddings. Utilizing cosine similarity, we matched the word embeddings to the user input, thus creating an intuitive web application for the recommendation system.
Inspiration behind the Project
I really love doing research, but finding the right professor to work with can be tough. I tried searching online, but it took forever, and the information wasn't always accurate. Eventually, I had to rely on a connection from one of my current professors to find the right match. This got me thinking – what if there was a better way? What if there was a way for people interested in research to easily find professors who share their interests? This could help us connect faster and even boost interest in research, which could improve literacy rates.
Tech Stack
Machine Learning Model:
Gemini
Bert
cosine_similarity
Frontend:
Streamlit - Python framework
Backend
Pandas
Scholarly
API's
sklearn
Hackathon Track
Graduate
Project Repo
https://github.com/shrutivarade/GuideGenie
Demo Link
https://guide-genie.streamlit.app
Demo Video/Photos
https://liveumb-my.sharepoint.com/:f:/g/personal/n_jain001_umb_edu/EkcK1OhCATRMm2zycONinpIBpgbDBZ9LLoD3so6L7tvJYw?e=gcCFsV
Anything Else?
No response
Rules and Code of Conduct
The text was updated successfully, but these errors were encountered: