# Analyze large and publicly available datasets ![alt text](https://files.slack.com/files-pri/T0HTW3H0V-F060Q793TNX/oue_034.png?pub_secret=1cf111f42b) prompt used: "Analyze large and publicly available datasets." ## description Students can use AI tools to analyze large, publicly available datasets as a way to develop a preliminary understanding of a complex dataset. AI tools like the advanced data analysis capability of GPT 4 can parse datasets and identify patterns, trends, etc. In addition to performing analyses, AI tools can also help students visualize the data. ## activity 1. For this activity, we're going to use OpenAI's GPT 4 advanced data analysis extension (rather than the HUIT Sandbox, since it doesn't currently have this capability). 2. Using the advanced data analysis extension in GPT 4, upload a large dataset of your choosing 3. ask chatGPT to analyze the data for you 4. ask chatGPT what this study is attempting to measure 5. ask chatGPT a follow-up question based on the results you're getting. --- ## Transcript of activity example: * USER: * We're going to use a dataset about mice that looks like this in csv form ![alt text](https://files.slack.com/files-pri/T0HTW3H0V-F062GJFGNUV/screen_shot_2023-10-24_at_11.45.22_am.png?pub_secret=e706662e9c) ```prompt: can you analyze this data for me?``` * AI: ![alt text](https://files.slack.com/files-pri/T0HTW3H0V-F063EGSFBPA/screen_shot_2023-10-25_at_10.09.14_am.png?pub_secret=af3ce30de8) ```prompt 2: what is this study attempting to measure?``` ![alt text](https://files.slack.com/files-pri/T0HTW3H0V-F062RE51YKD/screen_shot_2023-10-25_at_10.08.59_am.png?pub_secret=ae4f9c5733) ```prompt 3: Can you help me understand what this study has to do with Down Syndrome?``` ![alt text](https://files.slack.com/files-pri/T0HTW3H0V-F063EH6AU48/screen_shot_2023-10-25_at_10.11.15_am.png?pub_secret=d2aa785a7f)