With ChatGPT, you can do machine learning on your data without knowing how to code
In this video, I show how you can use ChatGPT and the Noteable plugin to use a machine learning model to classify your data without knowing how to code in Python.
Blog post:
www.chadskelton.com/2023/05/us...
Datasets:
github.com/chadskelton/data/t...
Пікірлер: 21
Chad it just so happens I went to Kwantlen college in Surrey. I recall how much better education was there at Kwantlen as opposed to Universities. You are doing a very fine job teaching your students this great technology. Congratulations on being such a pioneer. Kwantlen is blessed to have you as a professor.
Great work mate, I'm also an educator and an analyst and am looking at this stuff with a lot of excitement!
Love these videos! Thanks a lot, Chad!
That is good information Chad, I am following along with your test dataset. Thanks!
Another excellent video. Thank you, Chad!
like your video, very clear and clean way to explain thing, subscribed
You are the real MVP !
Hi Chad, I came across your videos a few weeks ago while awaiting a code interpreter, and I must say they've been quite informative. Thank you for taking the time to put these resources together. This particular video has inspired me to explore similar methods in my own work. It would be interesting to know, are you considering any projects with the Natural Language Toolkit in the future? Your insights would be greatly appreciated. Keep up the excellent work!
@chadskeltondata
Жыл бұрын
Was thinking about it. I don't know much about what NLTK can do aside from Entity Extraction so if you've got ideas for what might be worth trying, and what data or documents would be good as a test case, I'm all ears. :)
@jt1923
Жыл бұрын
@@chadskeltondata NLTK can offer fascinating insights for textual analysis, particularly for literary works. For instance, analyzing Shakespeare's works could make an intriguing project. Here's a general idea: -Choose a research question related to Shakespeare's work (e.g., linguistic patterns, metaphors in sonnets, role of female characters). -Gather the text data, available from sources like Project Gutenberg or the Folger Shakespeare Library. -Pre-process the data: tokenize, remove stopwords, convert to lowercase. -Apply appropriate NLP techniques such as text analysis (word frequencies, n-grams), sentiment analysis, topic modeling (themes in works), or stylometry (writing style). -Visualize the findings: word clouds, bar charts, network graphs etc. -Interpret the results in relation to the research question and discuss their significance. -Compile everything into a research paper format. You could showcase how NLTK can bring a new perspective to such classical texts and stimulate engaging conversations around Shakespeare's works. I'd love to see what you come up with! -JT
Hey Chad, thanks for this videos, I'm trying to see if it's possible through notable to connect chat gpt to my WordPress data base, I run an online academy for photography and managed to get interesting results with one csv, but would be much more useful if it could check really time data instead of having to export every time a csv. The data base is very large but we could point it out the important stuff or maybe it could figure on its own. Is it even possible at this time? The data base is sql
@chadskeltondata
Жыл бұрын
Not my area of expertise but, at least in theory, Noteable can connect to some databases as described in this video: kzread.info/dash/bejne/iIpsxtiioJm0Y9Y.html
@Jonhernandezeducation
Жыл бұрын
Thanks will take a look!
keep product more videooooooo
Chad Skeleton!! ☠
as a general strategy, would it be wise to ask to chat gpt to not just try the first option (the one leading the forest tree classification) but also the other options, asking which one performed better on the sample test?
@jt1923
Жыл бұрын
You're right. Having the correct categories beforehand aids in illustrating how machine learning operates. Although it does pose a challenge in understanding how to choose a model or proceed without such answers. Perhaps that's precisely why the video was presented in this way. This has certainly sparked my interest in experimenting with similar methods.
@chadskeltondata
Жыл бұрын
Good point. I might try this in future! :)
what is the data size noteable can handle, if we have data in lakhs of rows, will it be able to handle it, and open ai models have tokens limit, so how does it work in the backend to generate the insights based on overall data
@chadskeltondata
9 ай бұрын
Noteable can handle pretty large datasets. I don't know if there's a firm upper limit. Tokens don't affect ChatGPT as I'm using it. The limit on ChatGPT plus is how many questions you can ask it within a set period of time.
Can I use this for data analysis of Marketing campaigns and provide dashboard reports and insights for my SMMA clients? I dont know coding at all. Will it be possible with a couple of good prompts?