Mirrors/Data_ScienceUse_Cases
2
0
Fork 0
mirror of https://github.com/youronlydimwit/Data_ScienceUse_Cases.git synced 2026-01-03 11:38:10 +01:00
Code Issues Packages Projects Releases Wiki Activity

Update readme.md

Browse Source
This commit is contained in:
Sang Putu Sandhyana Yogi
2024-11-29 17:24:42 +07:00
committed by GitHub
parent 51d36cea7a
commit 73ce05418d
1 changed files with 18 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

23
Predictions/Random Forest/readme.md
View File

@@ -2,8 +2,21 @@
## Diabetes Prediction - `DONE`
- No EDA
- Using `RandomForestClassifier`, `SMOTE`, `sklearn`, `pandas` libraries
- Attained Accuracy of 75.97%
- Diabetes as Class `1` and No Diabetes as Class `0`
- Precision (Class 0) = **0.84**, Precision (Class 1) = **0.64**
- Recall (Class 0) = **0.77**, Recall (Class 1) = **0.75**
- F1-score (Class 0) = **0.80**, F1-score (Class 1) = **0.69**
- **Result:**
- Attained Accuracy of 75.97%
- Diabetes as Class `1` and No Diabetes as Class `0`
- Precision (Class 0) = **0.84**, Precision (Class 1) = **0.64**
- Recall (Class 0) = **0.77**, Recall (Class 1) = **0.75**
- F1-score (Class 0) = **0.80**, F1-score (Class 1) = **0.69**
## Digital Marketing Expense Prediction - `DONE`
- Performed Exploratory Data Analysis (EDA)
- Using `RandomForestRegressor`, `sklearn`, `seaborn`, `pandas`, `matplotlib` libraries
- Dataset contains features:
- `Impressions`, `Reach`, `Video_plays`, `Link_clicks`, `Engagement`, and `Live_time`
- Target variable: `Amount_spent` (in IDR)
- **Result:**
- Attained R² of **0.86**
- `live_time` contributes 80% to the prediction of `Amount_spent`
- Other features provide smaller but meaningful contributions
- The model demonstrates robust handling of non-linear relationships and multicollinearity.