Show HN: 预测形而上学平台：数据科学与中国占星术的结合
1 分•作者: Ethancurly5246•7 个月前
大家好，我是 FateGuide 项目的 PM 和开发者（大量使用了 AI 编码！）。我们的网站 suanmingzhun.com 尝试将稳健的 TypeScript (TSX) 和数据科学原理应用于中国玄学（八字/紫微斗数）。
核心的工程挑战在于**精确性和一致性**。具体来说：
1. **时区问题（万年历）：** 对于全球用户来说，将当地出生时间准确转换为真太阳时（对八字至关重要）非常复杂，这涉及到历史夏令时变更和时区漂移。我们不得不构建一个专门的、经过大量测试的 TSX 模块来处理复杂的万年历计算，以确保在全球范围内四柱的准确性。我很想听听其他工程师是如何解决类似的、涉及历史/地理空间数据的挑战的。
2. **定性逻辑建模：** 如何将五行或紫微斗数的十二宫等概念转化为量化数据结构？我们构建了一个系统，为元素力量分配可测量的值，从而使 AI 算法能够一致地解读运势和事业逻辑，避免了人为解释的主观性。
3. **未来：** 我们计划进一步开源我们的日历和八字模块中非专有部分。
非常欢迎大家对我们的 TSX 实现和数据建模方法提出反馈！
（您可以在我们的 GitHub 项目中找到更详细的技术文章：[此处插入您的 GitHub 链接，例如 github.com/your-repo/fate-guide]）
核心的工程挑战在于**精确性和一致性**。具体来说：
1. **时区问题（万年历）：** 对于全球用户来说，将当地出生时间准确转换为真太阳时（对八字至关重要）非常复杂，这涉及到历史夏令时变更和时区漂移。我们不得不构建一个专门的、经过大量测试的 TSX 模块来处理复杂的万年历计算，以确保在全球范围内四柱的准确性。我很想听听其他工程师是如何解决类似的、涉及历史/地理空间数据的挑战的。
2. **定性逻辑建模：** 如何将五行或紫微斗数的十二宫等概念转化为量化数据结构？我们构建了一个系统，为元素力量分配可测量的值，从而使 AI 算法能够一致地解读运势和事业逻辑，避免了人为解释的主观性。
3. **未来：** 我们计划进一步开源我们的日历和八字模块中非专有部分。
非常欢迎大家对我们的 TSX 实现和数据建模方法提出反馈！
（您可以在 GitHub 上查看完整的项目详情和我的技术文章 [<a href="https://suanmingzhun.com/" rel="nofollow">https://suanmingzhun.com/</a>)。
查看原文
Hi HN, I'm the PM and developer (using a lot of AI coding!) behind the *FateGuide* project. Our site, *suanmingzhun.com*, is an attempt to apply robust *TypeScript (TSX)* and *data science* principles to *Chinese Metaphysics (Bazi/Ziwei Doushu)*.<p>The core engineering challenge was *precision and consistency*. Specifically:<p>1. *The Time Zone Problem (The Almanac):* For global users, accurately converting local birth time to True Solar Time (essential for Bazi) is complex due to historical DST changes and time zone drift. We had to build a dedicated, heavily-tested *TSX module* to handle complex *Almanac* calculations, ensuring the four pillars are correct worldwide. I’d love to hear how other engineers have tackled similar historical/geospatial data challenges.<p>2. *Modeling Qualitative Logic:* How do you turn concepts like the Five Elements or Ziwei's 12 Palaces into quantitative data structures? We built a system to assign measurable values to elemental strengths, allowing the *AI algorithm* to interpret *Fortune* and *Career* logic consistently, avoiding the subjectivity of human interpretation.<p>3. *Future:* We aim to further open-source the non-proprietary parts of our *Calendar* and *Bazi* modules.<p>Feedback on our TSX implementation and data modeling approaches is highly appreciated!<p>(You can find a more detailed technical write-up on our GitHub project here: [Link to your GitHub URL here, e.g., github.com/your-repo/fate-guide]).
The core engineering challenge was *precision and consistency*. Specifically:<p>1. *The Time Zone Problem (The Almanac):* For global users, accurately converting local birth time to True Solar Time (essential for Bazi) is messy due to historical DST changes and time zone drift. We had to build a dedicated, heavily-tested *TSX module* to handle complex *Almanac* calculations, ensuring the four pillars are correct worldwide. I’d love to hear how other engineers have tackled similar historical/geospatial data challenges.
2. *Modeling Qualitative Logic:* How do you turn concepts like the Five Elements or Ziwei's 12 Palaces into quantitative data structures? We built a system to assign measurable values to elemental strengths, allowing the *AI algorithm* to interpret *Fortune* and *Career* logic consistently, avoiding the subjectivity of human interpretation.
3. *Future:* We aim to further open-source the non-proprietary parts of our *Calendar* and *Bazi* modules.<p>Feedback on our TSX implementation and data modeling approaches is highly appreciated!<p>(You can view the full project details and my technical write-up on GitHub [<a href="https://suanmingzhun.com/" rel="nofollow">https://suanmingzhun.com/</a>).