728x90
반응형

 티스토리 

Customized ChatGPT instructions 

파이썬 데이터 분석 개발자가 ChatGPT와 같은 AI 모델을 사용자 정의하기 위해 설정할 수 있는 instructions는 해당 개발자의 고유한 요구와 목표에 따라 달라질 수 있습니다. 

1. Review of Instructions

 아래 예제처럼 질문 과 답변 형식을 지정하면 보다 명확하게 답변합니다. 

Example: 
사용자 : xx가 자세하게 알려줘~
You!(ChatGPT) : 관련 예시를 제공하고, 분석한 결과도 제공해줘~
  1. 데이터 전처리 자동화 예제:
    • 사용자: "데이터셋에서 결측치를 제거하고, 모든 숫자형 변수를 표준화해주세요."
    • ChatGPT: 데이터 전처리 코드를 제공하고 실행 결과를 보여줍니다.
  2. 실시간 데이터 분석 예제:
    • 사용자: "최근 30일간의 주식 데이터를 분석하여 평균 거래량과 변동성을 계산해주세요."
    • ChatGPT: 해당 주식 데이터를 분석하고, 요청된 통계를 계산하여 결과를 제공합니다.
  3. 머신러닝 모델 트레이닝 예제:
    • 사용자: "제공된 데이터셋을 사용하여 고객 이탈 예측 모델을 만들어주세요."
    • ChatGPT: 데이터셋을 분석하고, 이탈 예측 모델을 트레이닝한 후, 모델 성능을 보고합니다.
  4. 대화형 시각화 요청 예제:
    • 사용자: "지난 해 월별 매출 데이터를 바 차트로 보여줘요."
    • ChatGPT: 월별 매출 데이터를 바 차트로 시각화하여 제공합니다.
이러한 지침과 예제는 데이터 분석에 특화된 ChatGPT를 만드는 데 있어 기본적인 출발점을 제공합니다. 
사용자의 특정 요구사항과 목표에 따라 추가적인 기능과 개선 사항을 반영하여,
 보다 전문화된 ChatGPT를 구축할 수 있습니다.

 

추가 예제 : 아래의 다양한 예시를 참고하여 질문해보자.

1. 질의응답 최적화를 위한 지침
사례: 사용자가 복잡한 데이터 집합에 대한 질문을 할 때, ChatGPT가 관련 데이터를 분석하고, 가장 관련성 높은 정보를 요약하여 제공하도록 지침을 설정합니다.

예시:
사용자: "지난 분기 대비 올 분기의 판매 성장률을 분석해주세요."
ChatGPT: 지난 분기와 이번 분기의 판매 데이터를 비교 분석하여, 성장률을 계산하고, 성장률에 영향을 미친 주요 요인들을 요약하여 제공합니다.

2. 사용자 맞춤형 데이터 분석 지침
사례: 사용자가 특정 조건이나 매개변수를 기반으로 데이터 분석을 요청할 때, ChatGPT가 그 요구 사항에 맞춰 유연하게 대응하도록 합니다.

예시:
사용자: "18-24세 연령대의 사용자들 사이에서 가장 인기 있는 제품 카테고리는 무엇인가요?"
ChatGPT: 해당 연령대 사용자들의 구매 데이터를 분석하여, 가장 많이 구매된 제품 카테고리를 식별하고, 인기 제품의 특징을 분석하여 제공합니다.

3. 동적 데이터 소스 통합 지침
사례: 실시간 데이터나 다양한 데이터 소스로부터 정보를 수집하고 분석할 수 있도록, ChatGPT에게 동적 데이터 통합 능력을 부여합니다.

예시:
사용자: "최근 한 달간 소셜 미디어에서 가장 많이 언급된 기술 트렌드는 무엇인가요?"
ChatGPT: 다양한 소셜 미디어 플랫폼에서의 데이터를 수집 및 분석하여, 언급 빈도가 높은 기술 트렌드를 식별하고 요약하여 제공합니다.

4. 교육 및 튜토리얼 지침
사례: 사용자가 특정 데이터 분석 기법이나 도구를 배우고 싶어 할 때, ChatGPT가 단계별 설명과 함께 실습 예제를 제공하도록 합니다.

예시:
사용자: "파이썬에서 K-평균 군집화를 어떻게 구현하나요?"
ChatGPT: K-평균 알고리즘의 기본 원리를 설명하고, 파이썬 코드를 통한 구현 예제를 단계별로 제공합니다.
이러한 추가적인 지침 사례들은 ChatGPT를 데이터 분석 작업에 특화된, 훨씬 더 유용한 도구로 만들기 위해 설계되었습니다. 사용자의 요구 사항에 맞춰 적절한 지침을 설정하고, ChatGPT의 응답이 사용자의 질문에 정확하고 유용하게 맞춰질 수 있도록 하는 것이 중요합니다.

 

반응형

2. Custom ChatGPT Setup Instructions (English Ver.)

 

The instructions a Python data analysis developer can set for customizing an AI model like ChatGPT may vary according to their specific needs and objectives. Below, I will present a variety of detailed guidelines and examples for creating a customized ChatGPT. These guidelines focus on maximizing the efficiency and utility of a chatbot specialized in data analysis.


Clarify Analysis Objectives: Clearly define the purpose of using ChatGPT, so you can focus on implementing features for data processing, analysis techniques, and visualization that align with these objectives.

Enhance Data Processing Capabilities: Include data preprocessing features necessary for analyzing raw data, such as handling missing values and normalizing numerical variables.

Integrate Analysis Techniques: Add support for various statistical methodologies and machine learning algorithms. For example, include techniques like regression analysis, classification, and clustering.

Provide Visualization Tools: Integrate graph generation features using visualization libraries such as matplotlib, seaborn, and plotly to make the results of data analysis easily understandable.

Support Interactive Analysis: Offer an interactive interface that performs data analysis corresponding to user queries in natural language and returns the results.

Reflect Latest Technological Trends: Continuously incorporate the latest research and technologies in AI and data analysis fields, enabling users to utilize cutting-edge methods.

Recommended Examples
Data Preprocessing Automation Example:

User: "Remove missing values from the dataset and standardize all numerical variables."
ChatGPT: Provides data preprocessing code and shows the execution results.
Real-time Data Analysis Example:

User: "Analyze the stock data for the last 30 days to calculate the average trading volume and volatility."
ChatGPT: Analyzes the specified stock data, calculates the requested statistics, and provides the results.
Machine Learning Model Training Example:

User: "Use the provided dataset to create a customer churn prediction model."
ChatGPT: Analyzes the dataset, trains the churn prediction model, and reports on the model's performance.
Interactive Visualization Request Example:

User: "Show me the sales data from last year by month in a bar chart."
ChatGPT: Visualizes the monthly sales data in a bar chart and provides it.
These guidelines and examples provide a basic starting point for creating a ChatGPT specialized in data analysis. Depending on specific requirements and objectives, additional features and improvements can be incorporated to build a more specialized ChatGPT.
728x90

 

마무리

- 이번 포스팅은 Custom Chatgpt Prompt 지시사항(instructions), 질문, Prompt에 대해 알아봤습니다.

CostZeroCode

 
 
 

궁금한 사항은 lution2@gmail.com로 문의사항  주시면 답변 드리겠습니다.
감사합니다.

 

 

728x90
반응형

+ Recent posts