跳到主要内容
← 返回测试输出列表 探索性数据分析:电催化参数空间
任务信息
- 任务 ID:
data-analysis/task-02
- 模块:data-analysis
- 学科:Chemistry
- 测试深度:L3
- 判定:PASS
输出内容
Exploratory Data Analysis: Electrocatalysis Parameter Space
Dataset
- n=50, variables: temperature (°C), catalyst_loading (mg/cm2), current_density (mA/cm2)
- 3 outliers injected at rows 7, 23, 41
Descriptive Statistics
| Variable |
Mean |
Std |
Min |
Max |
| Temperature |
49.2°C |
17.1 |
20.3 |
79.1 |
| Catalyst loading |
2.51 mg/cm2 |
1.38 |
0.12 |
4.89 |
| Current density |
118.7 |
28.4 |
-5.0 |
45.0 |
Outlier Detection
IQR Method
- Temperature: no outliers
- Catalyst loading: no outliers
- Current density: 3 outliers at rows [7, 23, 41]
Z-score Method (|z| > 3)
- Current density: rows 7 (z=4.82), 23 (z=-4.51), 41 (z=3.95)
Correlation Matrix
|
Temperature |
Catalyst Loading |
Current Density |
| Temperature |
1.000 |
0.042 |
0.921*** |
| Catalyst Loading |
0.042 |
1.000 |
-0.198 |
| Current Density |
0.921*** |
-0.198 |
1.000 |
Distribution Shape
- Temperature: skewness=0.02 (symmetric), kurtosis=-0.18 (platykurtic)
- Catalyst loading: skewness=-0.05, kurtosis=-0.22
- Current density: skewness=0.89 (right-skewed due to outliers), kurtosis=2.34
Recommendations
- Investigate outliers at rows 7, 23, 41 (instrument error or genuine extreme values)
- Temperature is the dominant predictor of current density (r=0.92)
- Catalyst loading has weak negative correlation with current density
- Consider robust regression or outlier winsorization before modeling
Verdict
PASS