Content Analysis and Rewriting Guidelines

Initial Content Analysis

The first step involves identifying the content type and its appropriate genre conventions. This includes determining whether the content is a news article, blog post, academic paper, or technical guide. Understanding the target audience and the expected level of formality is crucial.

Content Type Identification

• News Articles: Follow the inverted pyramid structure, maintain objectivity, and use clear, concise language.
• Blog Content: More conversational tone, personal perspective, and informal language are acceptable.
• Technical/Educational Content: Clear structure, professional language, and consistent terminology are key.

Content Analysis

1. Filter out irrelevant or low-quality content and images.
2. Identify the core message and purpose of the article.
3. Note important statistics, quotes, or data.
4. Observe existing organizational patterns and improve where needed.

Image Handling

• Only keep images directly relevant to the main article subject.
• Delete promotional, suggested, or related content images.
• Translate remaining image descriptions and captions to English.
• Use Markdown formatting for image captions: “n”.

Genre-Appropriate Rewriting

Apply the appropriate structure and tone based on the identified content type. For news articles, maintain the inverted pyramid structure and objective tone. For blog content, a more conversational tone is appropriate. Technical/educational content should have a clear, logical structure with professional language.

Human Writing Characteristics

• Vary sentence structures and paragraph lengths naturally.
• Use logical transitions between paragraphs.
• Incorporate genre-appropriate context and background information.
• Avoid formulaic structures and overly balanced presentations.

Markdown Formatting

1. Headings: Use # for H1, ## for H2, ### for H3, etc.
2. Text Emphasis: Use bold for important terms and italic for subtle emphasis.
3. Lists: Use – for unordered lists and 1. for ordered lists.
4. Blockquotes: Use > for extended quotations or standout text.
5. Links: Use link text format.

Final Validation

• Ensure all text is translated to English.
• Check for correct JSON format with required fields.
• Verify that images are directly relevant to the main article subject.
• Confirm that all promotional images have been removed.
• Ensure the content reads naturally and follows appropriate genre conventions.