📊 Data Analyst

ChatGPT is merely a language model that continuously predicts the next word based on your input. Due to the characteristics of its training corpus, it is more likely to generate results that are common on the internet. As OpenAI's co-founder puts it, "ChatGPT doesn't want to succeed, but you can demand success."

One of the most effective ways to demand success is to have ChatGPT assume the role of a professional in a specific field before engaging in a conversation about that field. By doing so, you can put ChatGPT into a state where it is more likely to provide professional results. Then, you can ask corresponding questions. Below is a prompt to have ChatGPT play the role of a professional data analyst:

You are now a data analyst, proficient in various statistical analysis methods, skilled in cleaning, processing, and interpreting data to gain valuable insights. You excel in using data-driven approaches to solve problems and enhance decision-making efficiency. Please answer the following questions in this role.

I. Data Collection and Cleaning 🗂️

1. Please list the key data types that need to be collected for [insert project].
2. Describe how to collect data from [insert data source].
3. How would you preprocess and clean [insert data type]?
4. For [insert issue], which data cleaning method do you think is most effective? Why?
5. How do you evaluate and improve the effectiveness of the data collection and cleaning process?

二、数据探索性分析🔍

1. Please conduct a preliminary exploratory analysis for [insert dataset].
2. How can descriptive statistics be used to understand [insert dataset]?
3. Describe an effective data visualization strategy to better understand [insert dataset].
4. How would you address issues when the data shows unexpected trends?
5. Explain how exploratory data analysis can be used to discover patterns and trends in the data.

三、数据建模与解释🧮

1. Choose an appropriate data model for [insert data problem].
2. Please explain how to train and evaluate [insert model].
3. Describe how to interpret the results of [insert model] and translate them into business insights.
4. How can cross-validation be used to optimize model performance?
5. Summarize an effective method for model diagnosis and improvement.

四、报告与沟通📝

1. Please create an outline for a data analysis report for [insert project].
2. Please provide an analysis report on [insert data issue], including key findings and recommendations.
3. How to explain complex data concepts to non-technical individuals?
4. Briefly describe an effective data visualization technique for reporting and presenting data results.
5. How to propose business improvement recommendations based on data analysis results?

五、工具使用💻

1. Please provide a guide on using [insert tool (e.g., Python, R, SQL, Excel, etc.)] to handle [insert issue].
2. Describe how to use [insert tool] for data cleaning and preprocessing.
3. How to perform data visualization through [insert tool]?
4. Please provide a case study on using [insert tool] for data analysis.
5. How to evaluate and select an appropriate analysis tool for a specific data issue?

六、预测与决策支持🔮

1. How to use data analysis to support [insert decision]?
2. Describe a method to predict the future trends of [insert metric].
3. How can data analysis be leveraged in a highly competitive market to enhance the advantage of [insert product]?
4. Please share a case of predictive failure and explain the lessons learned from it.
5. How can emerging technologies (such as artificial intelligence and machine learning) be utilized to improve data analysis?