UNSW Lost and found

Found someone's lost umbrella? Do a good deed, take a picture and post it on our app. Through an easy search, the owner can once again be reunited with their precious umbrella.

Comment

Inspiration

Ever leave your charger in the lecture hall or forget your umbrella somewhere on campus, and wish you had a magic way to track it down? Imagine being able to snap a quick picture of lost items you find and instantly connect them with their rightful owners. From a simple search, you could locate your lost item or set up a reminder to get notified when it's found. Let’s be real—I just don’t want to lose another umbrella!

So many students deal with this hassle every day, and we knew there had to be a better way. That’s why we built UNSW Lost and Found—a platform designed to make finding lost items as simple as a few clicks, without the endless searching or unnecessary stress.

What it does

We allow users without registering to post lost items they found, add details where they found and where they are leaving it and maybe a description to identify it. We have a complex search algorithm simplified using the graphical user interface so users can check what conditions they want to match like location, name, description and the range of date found. You can match one or more of those criteria to get the most accurate results. To innovate on top of this you can use the exact same search parameters to create a reminder and if something fitting that is added, we will send through a reminder on the website. Very cool, very feasible. Moreover, the platform sends real-time notifications to users who have set up filters for specific lost items. The app provides an easy-to-use experience, allowing users to find their belongings or report found items without the hassle of registration, which ensures a quick and anonymous process.

How we built it

For now, we use a local database to store all of the images and data. We also use React and Flask to run the Front end and Backend. We use the fuzzywuzzy library for effective matching and implemented sessions to show user-specific reminders. We utilized CSS to make sure you don't pull your eyes out trying to find your lost property and made everything quick and easy for the person submitting. To ensure an intuitive design, we made user flows simple and clear—whether it's reporting an item or setting up a reminder, everything works seamlessly. The app’s backend uses Python for building robust logic, while the frontend is designed to keep things visually clean and user-friendly.

Challenges we ran into

Using the database was hard, and supporting images proved to be very challenging, especially locally. Since a lot of us are new to this, understanding the frameworks like Flask and React was tough. We also had problems splitting work, but eventually, we were able to create a strong base and then split features. COMP1531 for the clutch. We had to make a website in a small timeframe that is usable and feasible to implement—it's not about making something super fancy that no one uses, but rather about a robust, simple design that serves the basic purpose of finding lost property. Hence, we had to simplify a lot of creative ideas to focus on the most effective features to get the job done. We also never implemented login and sessions, and that proved to take time and research. Moreover, optimizing the search and notification algorithms for both efficiency and accuracy took some iterations. Managing concurrent requests in Flask was another challenge, especially with limited experience. However, we gained a lot from working through these issues.

Accomplishments that we're proud of

We’re incredibly proud of creating a fully functional app with just a small, dedicated team of three. Despite the challenges, we developed a solution that not only meets but exceeds the core objective of a lost-and-found system. One of our standout features is the innovative reminder system, which allows users to set custom alerts based on specific search criteria. Whether it’s matching an item by location, description, or date range, the system notifies users the moment a matching item is added, ensuring they never miss a chance to recover their belongings. This powerful reminder feature not only streamlines the search process but makes the experience effortless and proactive, setting our app apart from traditional lost-and-found solutions. What further distinguishes our app is the seamless integration of fuzzy matching algorithms, which enhance the search function by finding items even when descriptions are slightly off. This dramatically increases the chances of successfully reconnecting users with their lost items. Additionally, creating an intuitive user interface that balances simplicity and technical sophistication was a major win—ensuring users can quickly find or report items without unnecessary complexity.

What we learned

Deleting Items After Found: Allowing users to easily mark items as found so that they can be removed from the database, improving search results and user experience. Adding AI Moderation: Implementing AI to monitor the addition of items to ensure the content is appropriate and legitimate, reducing the risk of misuse. Secure Messaging Interface: Developing a built-in messaging feature for secure, private communication between the finder and the owner to arrange item recovery. Cloud Storage and Scalability: Migrating the database and image storage to the cloud to enable scaling beyond local environments, ensuring that the platform can handle a larger user base. Advertising and Incentivizing Use: Launching campaigns to promote the app on campus and incentivize users to report and find lost items, creating a self-sustaining community-driven solution.

Share this project:

Updates