Prompts

Note: The following prompts were tested in ChatGPT Plus for Web. This is a paid version.

Summarising a Dataset

Make sure you modify the prompt below to suit the data you need to analyse you'd like to review. Before you run this prompt make sure you upload your data file. 

Act as an expert data scientist. Summarise the distribution of [specific variable] in the dataset, including any notable trends or patterns.

Visualising Data

Make sure you modify the prompt below to suit the data you need to visualise you'd like to review. Before you run this prompt make sure you upload your data file. 

Act as an expert data scientist. Generate a [type of graph] graph using the following variables: 
[list of specific variables and their axes in the graph].

Follow-up prompts (optional):

Changes the colours at ...
Label the axes as...
Add error bars to...


Explanation

The prompt gives AI a role (data scientist), a task (summarising the dataset or generating graphs), requirements (the variables), and instructions (make the graph or a list of statistics).

In these cases a dataset must be provided and it will extract the appropriate information given in the instructions. There are many types of data analysis that can be done using ChatGPT Plus, including: Descriptive Analysis, Exploratory Data Analysis (EDA), Predictive Analysis, Data Visualisation, Clustering and Segmentation, Time Series Analysis, Classification and Regression, Sentiment Analysis, Anomaly Detection and A/B Testing.

You can use follow up prompts to further customise your output, particularly in data visualisation.


Examples

The screenshots below demonstrate how the AI can generate a graphs for a given dataset. In this case it shows a bar graph and pie graph and shows some customisation options.

The screenshots demonstrate how the AI can generate a graphs for a given dataset. In this case it shows a bar graph and

The screenshots demonstrate how the AI can generate a graphs for a given dataset. In this case it shows a bar graph and

The screenshots demonstrate how the AI can generate a graphs for a given dataset. In this case it shows a bar graph and

Source: AI in Education, The University of Sydney, https://canvas.sydney.edu.au/courses/51655/pages/conducting-exploratory-data-analysis?module_item_id=2412397
Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 License.

Back to '3.5: Transforming Your Work with Generative AI Tools'