The most pressing problem with existing green spots, is that they are underutilized. Relying on people to seek out green spots can lead to them being forgotten, and the money and effort wasted.
With our proposal, we will bring plant life to where the students already are, encourage interaction with the plants, provide ease of care with a monitoring app, and use machine learning to analyze sentiment of the students for future improvements.
Our vertical smart-planters will be wall mountable, and sized to fit in existing study spaces. They will have their own UV light supply, water, temperature, and humidity sensors. They will send data to a raspberry-pi, which can transfer files to an app developed for plant management.
When the plants need water, an indicator light will be activated on the pot. With water conveniently placed nearby, students will be encouraged to water the thirsty plant themselves. This promotes interaction with our plants, and can be a short healthy distraction.
If no one takes it upon themselves to water the plant, it will be watered automatically with simple irrigation.
After one of these spaces is built, we use sentiment analysis on social media posts to gauge student’s opinions. A simple web crawler can pull social media posts, filtering for location and key words.
Once the contents of these posts have been pulled, it can be processed by a language model that can determine the sentiment of the post. This will provide invaluable feedback on the effectiveness of our project at a glance, and will aid in making future investments. This will also allow CMU to detect under-utilized or disliked spaces quickly, without the need for surveying.
Built With
- 3d-printing
- arduino
- dall-e
- huggingface
- raspberry-pi
- roberta
Log in or sign up for Devpost to join the conversation.