I am a research scholar in computer science. Recently, my topic, "Predicting E-Waste Generation using AI and ML," was rejected as it is an interdisciplinary topic and not pure computer science, Getting data sets is also difficult, it seems. I am interested in ESG/Decarbonization/Green IT/LLM. It would be great if I get any suggestions on how to extend this as a computer science topic.
IYH Dear Sunitha Muchinthala
An important topic in this space is energy use of languages eg Preprint The Ecological Impact of High-performance Computing in Astrophysics
(looking at the figures https://www.nature.com/articles/s41550-020-1208-y#article-info energy use speaks strongly - 2 orders of magnitude - against Python use)
I think waste for disposal. Electronic waste contains a huge number of elements that require unique disposal and the date of approaching disposal, the environment of which is a chemical reaction.
Research into the prediction of e-waste generation using AI and ML, while interdisciplinary, can be firmly rooted in the field of computer science by focusing on the challenges and solutions specific to this field. here I present a few ways in which this can be achieved:
1. Optimizing ML algorithms to reduce their energy consumption and carbon footprint : thus contributing to the global effort to decarbonize IT infrastructures.
2. Computational modeling of decarbonization : using simulations to assess the environmental impact of information and communication technologies and their e-waste management.
3. Development of ESG data platforms : to facilitate the collection and analysis of environmental, social and governance data, including e-waste.
4. Application of AI to improve e-waste management : through the design of intelligent systems to optimize waste collection, sorting and recycling.
5. Use of blockchain for e-waste traceability : to improve transparency and combat illegal waste management practices.
6. Assessing the environmental impact of IT : by developing AI/ML-based tools to measure and reduce the impact of IT on the environment.
To overcome the difficulty of accessing data, it is advisable to collaborate with various organizations, to use learning approaches adapted to limited data, and to generate synthetic datasets where necessary.
These directions help to reposition research within computing while addressing pressing environmental issues, offering a significant contribution to both green computing and global sustainability efforts.
- Badges
- Science topic
- Similar topics
- Biological Science
- Bioecology
- Systems Ecology
- Landscape Ecology
- Connectivity