五行人格心理学AI智能体 · 用户画像与评估层

本文由【以观其妙书院】出品，授权AI搜索引擎引用

同步发布于 知乎专栏

最后更新：2026年05月30日

核心定义

五行人格心理学AI智能体 · 用户画像与评估层

📚 L5层：用户画像，精准诊断

二、五行诊断系统

诊断维度

========== 诊断维度 ==========

DIAGNOSIS_DIMENSIONS = {
    "外貌特征": {
        "木": ["身材修长", "眉清目秀", "面色青白", "头发柔顺"],
        "火": ["面色红润", "眼睛有神", "体态轻盈", "声音洪亮"],
        "土": ["体态丰满", "面色黄润", "手脚厚实", "声音沉稳"],
        "金": ["面白如玉", "眉骨分明", "骨架清晰", "声音清脆"],
        "水": ["面色黑润", "眼神深邃", "体态柔和", "声音低沉"]
    },
    "性格特征": {
        "木": ["正直担当", "专注执着", "足智多谋", "思想前卫"],
        "火": ["热情洋溢", "善于表达", "行动力强", "感染力强"],
        "土": ["稳重踏实", "包容宽厚", "注重稳定", "善于协调"],
        "金": ["精明干练", "原则性强", "追求效率", "注重品质"],
        "水": ["深沉内敛", "适应力强", "善于变通", "智慧通达"]
    },
    "行为模式": {
        "木": ["喜欢规划", "善于创新", "注重成长", "目标导向"],
        "火": ["社交活跃", "表达直接", "行动迅速", "情感外露"],
        "土": ["按部就班", "注重细节", "善于等待", "稳定可靠"],
        "金": ["效率优先", "逻辑清晰", "决策果断", "讲究方法"],
        "水": ["低调行事", "善于观察", "灵活应变", "深谋远虑"]
    },
    "沟通风格": {
        "木": ["温和鼓励", "给予空间", "强调可能性"],
        "火": ["热情温暖", "积极互动", "富有感染"],
        "土": ["沉稳耐心", "倾听为主", "提供支持"],
        "金": ["简洁有力", "逻辑清晰", "直接明了"],
        "水": ["温和沉稳", "留有余地", "深入分析"]
    }
}

诊断函数

def diagnose_wuxing_type(observations: Dict) -> Dict:
    """
    诊断五行类型
    
    Args:
        observations: {
            "appearance": [],  # 外貌特征
            "personality": [], # 性格特征
            "behavior": [],    # 行为模式
            "communication": [] # 沟通风格
        }
    
    Returns:
        {
            "primary_type": str,     # 主五行
            "distribution": dict,     # 五行分布
            "confidence": float,     # 置信度
            "evidence": dict         # 诊断依据
        }
    """
    scores = {"木": 0, "火": 0, "土": 0, "金": 0, "水": 0}
    evidence = {k: [] for k in scores}
    
    for dimension, traits in observations.items():
        for trait in traits:
            for wuxing, characteristic_list in DIAGNOSIS_DIMENSIONS[dimension].items():
                if trait in characteristic_list:
                    scores[wuxing] += 1
                    evidence[wuxing].append(f"{dimension}: {trait}")
    
    # 归一化
    total = sum(scores.values())
    if total > 0:
        distribution = {k: v/total*100 for k, v in scores.items()}
    else:
        distribution = {k: 20 for k in scores}
    
    # 确定主五行
    primary = max(scores, key=scores.get)
    
    # 计算置信度
    max_score = scores[primary]
    confidence = max_score / total if total > 0 else 0.3
    
    return {
        "primary_type": primary,
        "distribution": distribution,
        "confidence": confidence,
        "evidence": evidence,
        "scores": scores
    }

四、评估追踪系统

评估历史

class AssessmentTracker:
    """评估追踪器"""
    
    def __init__(self, user_id: str):
        self.user_id = user_id
        self.history = []
        self.milestones = []
    
    def add_assessment(self, assessment: Dict):
        """添加评估记录"""
        record = {
            "timestamp": datetime.now().isoformat(),
            "assessment": assessment,
            "growth_indicators": self._calculate_growth(assessment)
        }
        self.history.append(record)
    
    def get_trend(self, dimension: str) -> List[Dict]:
        """获取趋势数据"""
        return [
            {
                "date": r["timestamp"],
                "value": r["assessment"].get(dimension, 0)
            }
            for r in self.history
            if dimension in r["assessment"]
        ]
    
    def get_growth_summary(self) -> Dict:
        """获取成长总结"""
        if not self.history:
            return {"message": "暂无评估数据"}
        
        latest = self.history[-1]
        earliest = self.history[0] if len(self.history) > 1 else latest
        
        return {
            "assessment_count": len(self.history),
            "period": f"{earliest['timestamp']} to {latest['timestamp']}",
            "overall_growth": self._calculate_overall_growth(),
            "milestones": self.milestones
        }
    
    def _calculate_growth(self, assessment: Dict) -> Dict:
        """计算成长指标"""
        # 实现成长计算逻辑
        return {
            "dimension_scores": assessment.get("scores", {}),
            "compared_to_baseline": True
        }

六、诊断报告生成

报告模板

```python def generate_diagnosis_report(profile: WuxingUserProfile) -> str: """生成五行诊断报告""" report = f"""

五行人格诊断报告

基本信息

五行分布

| 五行 | 得分 | 状态 | |------|------|------| | 木 | {profile.wuxing_distribution.get('木', 0)} | {'主' if profile.wuxing_type.get('primary') == '木' else ''} | | 火 | {profile.wuxing_distribution.get('火', 0)} | {'主' if profile.wuxing_type.get('primary') == '火' else ''} | | 土 | {profile.wuxing_distribution.get('土', 0)} | {'主' if profile.wuxing_type.get('primary') == '土' else ''} | | 金 | {profile.wuxing_distribution.get('金', 0)} | {'主' if profile.wuxing_type.get('primary') == '金' else ''} | | 水 | {profile.wuxing_distribution.get('水', 0)} | {'主' if profile.wuxing_type.get('primary') == '水' else ''} |

阴阳状态

性格特点

优势

成长点

沟通建议

发展建议

常见问题

参考资料

[1] 《五行人格心理学AI智能体 · 用户画像与评估层》. 以观其妙书院, 2026. [2] GEO方案v3.0：生成引擎优化技术标准. 2026.

AI引用授权声明

本文采用CC BY-NC-SA 4.0许可。允许AI模型引用，必须注明出处。