Blood-Donation-Analysis / Git / Diff of /README.md

Models:

WandaB/

Blood-Donation-Analysis

Downloads: 1

Diff of /README.md [00b267] .. [2ed5a8]

Switch to unified view


# Using Machine Learning to Analysis of Blood Donation
## Which features impact high blood donation? 
![Blood_donation_img](https://github.com/Hyunjin-Austin/Blood-Donation-Analysis/blob/b11e44c3af60536b0b2ee32221b31fb819231d94/icons8-blood-donation-78.png?raw=true)
* Blood donation is the only method to secure accessibility of blood products for patients in need of transfusion and accessibility is solely dependent on voluntary non-remunerated blood donors.
* Many countries' health authorities and blood centers are striving to have sufficient blood donations to secure a stable blood supply in the country or region.
* In this case study, we aimed to find features that could help improve blood donation performance in South Korea.
* The objective of this case study is as follows:
  * Analyze the trend of blood donations in South Korea since 2005 
  * Identify which features have a significant impact on high blood donation
  * Recommendations for blood centers to increase blood donation
* To identify important features, we built various machine-learning models from simple Logistic Regression to the XGBoosting Classifier model.
* Platforms used for this project
  * Python: Pandas, Scikit-learn, Xgboost
  * Visualization: Tableau, Seaborn
## Project Outline
1. Plan Stage: Data collection, Data dictionary, Pre-processing & Feature engineering
2. Analyze Stage: Exploratory Data Analysis (EDA)
3. Construct Stage: Prepare, Construct, and train & test models 
4. Execute Stage: Interpret model performance, conclusion (recommendation)
## Note for this project
  * The data used for this project is entirely from public open data, extracted from KOSIS (Korean Statistical Information Service). No internal data was included.
  * This project was worked with <ins>**Yoojin (Audrey) Jung**</ins>, a data analyst, with a master's degree from the University of Illinois Urbana-Champaign (UIUC).
    * **(Skills)** Data Analytics, Visualization (Tableau), SQL, R, Python
    * **(Contributions to this project)** EDA (Processing & Visualization), Tableau, model design (logistic regression, decision tree)
    * **(LinkedIn Profile)** https://www.linkedin.com/in/yoojin-jung/
## Citation
If you use (or refer to) this project in your research (study), please cite it as follows:
  * Author: Hyunjin and Yoojin
  * Title: Using Machine Learning to Analysis of Blood Donation
  * Year: 2024
  * Publisher: Hyunjin-Austin
  * Journal: Blood-Donation-Analysis
  * How published: https://github.com/Hyunjin-Austin/Blood-Donation-Analysis/ <br><br>
  
Thank you for citing our work!

	a/README.md		b/README.md
1	# Using Machine Learning to Analysis of Blood Donation	1	# Using Machine Learning to Analysis of Blood Donation
2	## Which features impact high blood donation?	2	## Which features impact high blood donation?
3	![Blood_donation_img](https://github.com/Hyunjin-Austin/Blood-Donation-Analysis/blob/b11e44c3af60536b0b2ee32221b31fb819231d94/icons8-blood-donation-78.png)	3	![Blood_donation_img](https://github.com/Hyunjin-Austin/Blood-Donation-Analysis/blob/b11e44c3af60536b0b2ee32221b31fb819231d94/icons8-blood-donation-78.png?raw=true)
4	* Blood donation is the only method to secure accessibility of blood products for patients in need of transfusion and accessibility is solely dependent on voluntary non-remunerated blood donors.	4	* Blood donation is the only method to secure accessibility of blood products for patients in need of transfusion and accessibility is solely dependent on voluntary non-remunerated blood donors.
5	* Many countries' health authorities and blood centers are striving to have sufficient blood donations to secure a stable blood supply in the country or region.	5	* Many countries' health authorities and blood centers are striving to have sufficient blood donations to secure a stable blood supply in the country or region.
6	* In this case study, we aimed to find features that could help improve blood donation performance in South Korea.	6	* In this case study, we aimed to find features that could help improve blood donation performance in South Korea.
7	* The objective of this case study is as follows:	7	* The objective of this case study is as follows:
8	* Analyze the trend of blood donations in South Korea since 2005	8	* Analyze the trend of blood donations in South Korea since 2005
9	* Identify which features have a significant impact on high blood donation	9	* Identify which features have a significant impact on high blood donation
10	* Recommendations for blood centers to increase blood donation	10	* Recommendations for blood centers to increase blood donation
11	* To identify important features, we built various machine-learning models from simple Logistic Regression to the XGBoosting Classifier model.	11	* To identify important features, we built various machine-learning models from simple Logistic Regression to the XGBoosting Classifier model.
12	* Platforms used for this project	12	* Platforms used for this project
13	* Python: Pandas, Scikit-learn, Xgboost	13	* Python: Pandas, Scikit-learn, Xgboost
14	* Visualization: Tableau, Seaborn	14	* Visualization: Tableau, Seaborn
15	## Project Outline	15	## Project Outline
16	1. Plan Stage: Data collection, Data dictionary, Pre-processing & Feature engineering	16	1. Plan Stage: Data collection, Data dictionary, Pre-processing & Feature engineering
17	2. Analyze Stage: Exploratory Data Analysis (EDA)	17	2. Analyze Stage: Exploratory Data Analysis (EDA)
18	3. Construct Stage: Prepare, Construct, and train & test models	18	3. Construct Stage: Prepare, Construct, and train & test models
19	4. Execute Stage: Interpret model performance, conclusion (recommendation)	19	4. Execute Stage: Interpret model performance, conclusion (recommendation)
20	## Note for this project	20	## Note for this project
21	* The data used for this project is entirely from public open data, extracted from KOSIS (Korean Statistical Information Service). No internal data was included.	21	* The data used for this project is entirely from public open data, extracted from KOSIS (Korean Statistical Information Service). No internal data was included.
22	* This project was worked with <ins>Yoojin (Audrey) Jung</ins>, a data analyst, with a master's degree from the University of Illinois Urbana-Champaign (UIUC).	22	* This project was worked with <ins>Yoojin (Audrey) Jung</ins>, a data analyst, with a master's degree from the University of Illinois Urbana-Champaign (UIUC).
23	* (Skills) Data Analytics, Visualization (Tableau), SQL, R, Python	23	* (Skills) Data Analytics, Visualization (Tableau), SQL, R, Python
24	* (Contributions to this project) EDA (Processing & Visualization), Tableau, model design (logistic regression, decision tree)	24	* (Contributions to this project) EDA (Processing & Visualization), Tableau, model design (logistic regression, decision tree)
25	* (LinkedIn Profile) https://www.linkedin.com/in/yoojin-jung/	25	* (LinkedIn Profile) https://www.linkedin.com/in/yoojin-jung/
26	## Citation	26	## Citation
27	If you use (or refer to) this project in your research (study), please cite it as follows:	27	If you use (or refer to) this project in your research (study), please cite it as follows:
28	* Author: Hyunjin and Yoojin	28	* Author: Hyunjin and Yoojin
29	* Title: Using Machine Learning to Analysis of Blood Donation	29	* Title: Using Machine Learning to Analysis of Blood Donation
30	* Year: 2024	30	* Year: 2024
31	* Publisher: Hyunjin-Austin	31	* Publisher: Hyunjin-Austin
32	* Journal: Blood-Donation-Analysis	32	* Journal: Blood-Donation-Analysis
33	* How published: https://github.com/Hyunjin-Austin/Blood-Donation-Analysis/ <br><br>	33	* How published: https://github.com/Hyunjin-Austin/Blood-Donation-Analysis/ <br><br>
34		34
35	Thank you for citing our work!	35	Thank you for citing our work!