Family Fun Finder

Inspiration

We love hanging out with our families so we developed an app to help people discover enjoyable family-friendly destinations which may be time-consuming when searching on Google since current search results are often outdated & irrelevant.

What it does

Recommends a list of places for the family based on rating, # of reviews, & city

How we built it

We retrieved the data by scraping Google Maps then cleaned it to remove any inconsistencies, errors, or irrelevant information with Python (using Pandas) and Google Sheets. Next, we conducted feature extraction utilizing scikit-learn's TfidfVectorizer to convert text data into numerical representations. To build the model, we used scikit-learn's cosine_similarity then assessed our model's performance using evaluation metrics such as Precision & Recall. Lastly, we deployed & hosted our model on Streamlit so users can input specific parameters to receive personalized family-friendly places to visit.

Challenges we ran into

• We encountered some technical issues (handling columns with varying datatypes) while building & deploying & needed to go through many iterations for the desired output.

Accomplishments that we're proud of

• We're proud of diving into deploying an ML model on Streamlit, despite having limited knowledge in this area.

What we learned

• The process of deploying an ML model on Streamlit
• Best practices for UX/UI for designing a deployed ML model

What's next for Family Fun Finder

• Stay tuned to find out! ;)

Built With

• google-colab
• google-spreadsheets
• numpy
• pandas
• python
• scikit-learn
• streamlit
• vscode