Editing AI Prompt Engineering in Mandarin (section)

== Format Conversion ==
=== Convert the response to json ===
* 目的：將機器人的結果可以更結構化，減少後續資料清理的功
* 提示：「... And put your response into a JSON with keys "aaa", "bbb", "ccc":」、「分析底下這句話，直接以 JSON 回傳抓取到的資訊 ，包含 ...」<ref>[https://www.facebook.com/DotNetWalker/posts/pfbid0X6Hs5U3j3Hd6v7ddcgunUHwzAxpZTNK3NjRt9cSE5v3FPf38QGk4Lxg3AFJPtBWkl ‧NET Walker 大內行者 (2023). 既然 ChatGPT 那麼厲害，都聽得懂人話，那分析用戶的 intent 與 entities 不正是他的強項嗎?]</ref> 其他提示寫法「Provide them in JSON format with the following keys: book_id, title, author, genre.」<ref>[https://www.deeplearning.ai/short-courses/chatgpt-prompt-engineering-for-developers/ ChatGPT Prompt Engineering for Developers - DeepLearning.AI]、「Output a json object that contains the following keys: french_summary, num_names.」</ref>

修改前

<pre>
Human: Write 10 keywords, sentiment and Named entity recognition for following article: 
今天台北天氣真好，希望可以不用上班
AI: Keywords: Taipei, Weather, Work, Today. Sentiment: Positive. Named Entity Recognition: Taipei, Work.
</pre>

修改後

<pre>
Human: Write 10 keywords, sentiment and Named entity recognition for following article. And put your response into a JSON with keys "keywords", "sentiment", "entity":

今天台北天氣真好，希望可以不用上班
AI:
{
    "keywords": ["Taipei", "weather", "not working", "today"], 
    "sentiment": "positive", 
    "entity": ["Taipei", "today"]
}
</pre>

<pre>
AI: I am an AI created by OpenAI. How can I help you today?
Human: 分析底下這句話，直接以 JSON 回傳抓取到的資訊 ，包含 出發站（Departure）、到達站(arrival）、票種（TickeType）、數量 (Amount）、出發時間 (DepartureTime）。•我爱從台北到高雄的自由座三張，下午1:30出發。還有爾張優待票，去合中的，2:30出發
AI: { "Departure": "Taipei",  "Arrival": "Kaohsiung",  "TicketType": "Freedom Seats", "Amount": 3, "DepartureTime": "1:30 PM" }, { "TicketType": "Discounted", "Departure": "Hecheng", "DepartureTime": "2:30 PM" }
</pre>

=== Structure the Reply in Nested JSON Format ===

Desired format:
<pre>
{
  "output": {
    "keywords": ["keyword1", "keyword2", "keyword3", "..."],
    "summary": "A concise overview that captures the key elements and core ideas of the subject."
  }
}
</pre>

The prompt:
<pre>
Organize the response in a JSON structure. Use the primary key "output" that encompasses two nested keys: "keywords" and "points". Additionally, include a "summary" under the "output" key.
</pre>



=== Convert a transcript into article ===
提示文字的效果
* 改成通順的文章：將文章改成通順的文章，並加上標點符號
* 分段：適當地分段

將採訪逐字稿，轉成通順的文章
* 「請將文章改成通順與分段的文章：{{kbd | key=接上文章}}」
* 或「請整理成分段且通順的文章，並在每個段落加上黑體的副標題：{{kbd | key=接上逐字稿}}」
* 或 Convert the transcript into article: {{kbd | key=接上文章}}
* 或 Convert the transcript into a blog article: {{kbd | key=接上文章}}
* 或 幫我整理成通順的文章: {{kbd | key=接上文章}}<ref>[https://www.facebook.com/groups/2152027081656284/?multi_permalinks=2153876181471374&hoisted_section_header_type=recently_seen ChatGPT 生活運用 | Facebook]</ref> 


將採訪逐字稿，轉成通順的文章（加上標點符號），並且摘要重點
* 「summarize the transcript into a fluent article:{{kbd | key=接上文章}}」

=== Convert to table ===
Prompt: 
* "Convert to table: {{kbd | key=接上文章}}"<ref>[https://www.facebook.com/groups/aicomunity/posts/3450445248608199/ ChatGPT Experts | Nice trick "Convert to table": | Facebook]</ref>
* 或「請將下面文章整理成表格，欄位是 (1)  {{kbd | key=欄位}}： {{kbd | key=接上文章}}」