How to Anonymize Bazi Cases
Minimize direct and indirect identifiers, choose deletion, generalization, date shifting, aggregation, or synthetic cases, and disclose analytic changes.
Overview
Removing a name is not enough to anonymize a Bazi case. Exact birth date, time, place, chart, rare events, occupation, family structure, screenshots, and publication context can identify a person when combined. Start with data minimization and a documented re-identification review. Delete unnecessary fields, generalize events, shift or withhold dates, aggregate details, or create a clearly labeled synthetic case. If changing birth data changes the chart, do not present the result as the original person’s exact case. Pseudonyms and consent reduce some risk but do not guarantee anonymity.
At a glance
- Direct identifiers
- Name, username, contact, account, image, record number
- Quasi-identifiers
- Exact birth data, place, chart, rare events, job, family, publication context
- Controls
- Delete, generalize, shift, aggregate, withhold, or synthesize
- Analytic integrity
- Disclose when changes alter pillars, timing, or the claim
- Residual risk
- Pseudonyms and consent do not guarantee anonymity
- Review
- Test linkage risk before publication and after added context
Inventory identifiers and linkage paths
List direct identifiers and combinations of birth data, chart details, events, occupation, geography, family, screenshots, and external posts.
Assess what a motivated reader could link using public information, not only what the article explicitly names.
- Review combinations, not fields alone.
- Include visual and contextual clues.
Keep only data necessary for the teaching point
Remove unrelated events, exact dates, locations, family details, and quotes. Generalize or aggregate remaining context where the analysis permits.
A pseudonym is not anonymization when the exact chart and rare biography remain searchable.
- Minimize before transforming.
- Do not rely on a false name.
Choose transformations without falsifying the case
Use date shifting, event ranges, withheld pillars, composites, or synthetic examples according to the teaching need and disclose the method.
If altered birth data produces a different chart, label the example synthetic or reconstructed; never claim it is the person’s exact reading.
- Preserve analytic validity.
- Label synthetic and composite cases.
Reassess residual risk before release
Review consent scope, audience, search visibility, rare detail combinations, minors, third parties, and whether later articles could enable linkage.
Withhold the case when teaching value does not justify the residual risk. Support correction, deindexing, or withdrawal where appropriate.
- Use a publication stop rule.
- Recheck after contextual changes.
Sources and editorial basis
- 《淵海子平》 (Yuanhai Ziping)《淵海子平》的四柱、六親、格局及歲運細節說明命盤與罕見傳記組合可形成間接識別，匿名化不能只移除姓名A foundational Ziping reference for branch months, hidden stems, roots, and stem-branch relationships.
- 《三命通會》 (Sanming Tonghui)《三命通會》的精確干支、時辰、歲運與案例式材料用於評估日期位移、隱藏命盤或合成案例對分析完整性的影響A historical compilation covering stems and branches, Ten Gods, Nayin, Shen Sha, and multiple classical pattern systems.
- OpenFate Editorial Methodology案例匿名化頁盤點直接及準識別資訊，要求資料最小化、連結風險審查、轉換揭露及殘餘風險停止發布規則OpenFate separates deterministic chart calculation, traditional interpretation, and modern editorial explanation.