General principle
This project demands two main parts:
Firstly, the smart bin which identifies the code of the householder, and recognizes the volumen, the weight, the different kinds of garbage such as organic, plastic, metal, etc. This recognition is based on artificial intelligence. Thus, it uses advance algorithms trained by computer.
Second, the website and app from recyclebank works with a database from all householders that previously they have registered via online. The process of earning points is based on the amount of garbage and the kind of trash that people recycle in the smart bins.
Firstly, the smart bin which identifies the code of the householder, and recognizes the volumen, the weight, the different kinds of garbage such as organic, plastic, metal, etc. This recognition is based on artificial intelligence. Thus, it uses advance algorithms trained by computer.
Second, the website and app from recyclebank works with a database from all householders that previously they have registered via online. The process of earning points is based on the amount of garbage and the kind of trash that people recycle in the smart bins.
Technical overview - AI version
- Is it based on supervised or unsupervised learning? Recyclebank is based on supervised learning because it needs to train with new different types of garbage everyday.
- What kind of data are manipulated: values, images, sounds, words? Recyclebank manipulates images, sounds, shapes, colors, and weight from different types of garbage.
- What tasks does the algorithm have to perform? Classification, generation, pattern identification, clustering, regression, etc. The algorithm must identify the different types of garbage and then classify each of them in order to calculate the total score of points.
- Is it using datasets and how? What do these datasets contain, are data labelled, and if yes how? Where do the datasets come from? It uses a database to recognize previous materials, objects, or trash in order to make fast recognitions.The database is stored in the cloud of firebase, which lets the bins extract via online.
- What are the types of algorithms used? In which language are they programmed, with which librairies? This algorithm may use Javascript language with the library hellom5.js which lets you construct models of classification.
- Are there some potential biases we have to be careful of? This technology may have errors in the recognition process because there are different types of garbage that are new in the market or some materials that the algorithm never tested before.
Added value thanks to Artificial Intelligence
Based on artificial intelligence, the recyclebank may improve their system thanks to some features such as gps or geolocation which lets them connect with other neighbors or citizens from the community, and get more feedback from new householders in order to obtain more data to train and test the algorithm and improve it.
Another way to improve this project is feeding the algorithm with information about the information of shoppings or buyings, interactions products or services from the users or customers in order to find new attractive local business to attract new users.