Today my idea was try fantastic ChatGPT 4 (this is subscription version) in the data analytics (also, in the ChatGPT 4 is added special plugin 'Data analytics').
Topic which I am currently focusing on is: consumed electrical energy by type of energy source: fossil.
There is the example of prompt which I wanted to generated on the first time.
#My query: ‘Please generate analysis of this file (please use the same time range 2018-2021 - please skip 2022 year because data is not completed.) Please show top 5 countries which consumed the most energy from fossil.’
Comments: After send the csv file to analyzing I chosen only the countries (removed all huge regions, continents etc.) and asked GTP to focusing only for the common time range.
***
GPT answers: The top 5 countries that consumed the most energy from fossil fuels during the years 2018 to 2021 are:
- China: 20,896.62 TWh
- United States: 10,193.80 TWh
- India: 5,089.88 TWh
- Japan: 2,913.52 TWh
- Russia: 2,622.34 TWh
These figures represent the total electricity generated from fossil fuels in each of these countries across the four-year period.
Also I asked for the generating the columns chart of this prompt.
Ok, here's a short analysis of what GPT chat can do and can help us, of course we can ask it for further analysis:
- we can modify the query and ask for ready-made conclusion
- we can ask for visualizations
- we can ask for the Python script which is a basis of this analysis for the self-learning.
***
By the way, the topic of electricity and the types of its sources is an interesting topic that I would like to explore in the form of a larger project in the future.
The database which I used the generate a prompt comes from Our World in Data.
Source of data: https://ourworldindata.org/grapher/elec-mix-bar?tab=table&time=2018
Add comment
Comments