Data Analyst | Excel, SQL, Power BI | Designing dashboards, analyzing datasets, and reporting trends… Music (Ain’t No Stoppin Em).. #UFC. Follow & ifb.
BUILDING A “Heart failure prediction model” in PUBLIC 2
Good day #Datafam,
This is where the machine learning model got more complicated and also the areas I learned that context and domain knowledge is very essential to machine learning project.
#python
https://t.co/ZLXysokpqb
BUILDING A “Heart failure prediction model” in PUBLIC1
Good morning #Datafam
Heart disease is one of the leading causes of death worldwide.
Last week, my team wrapped up our capstone project building a machine learning system to predict heart failure from patient health records.
Dear Nigerians, PAY ATTENTION!
Lere Olayinka has deleted the tweet that exposed his crime. He deleted it without accountability. The DSS are yet to pick him up.
@OlayinkaLere is yet to explain to all Nigerians how he was able to gain unfettered access to INEC’s restricted area of the backend.
Wike’s aide is yet to tell us how he managed to publish Emeka Ike’s sensitive data. Professor Joash Amupitan has explanations to make as well. It will cost you nothing to retweet this, until INEC gives a satisfactory explanation.
Lere must tell us what happened to the IREV.
INEC wrote a whole load of rubbish, but failed to mention @OlayinkaLere name; not once!
WHAT DOES THIS TELL YOU?
INEC & whoever wrote the rubbish on their behalf, including Lere Olayinka, are laughing at Nigerians behind the scene. Any result/results declared by Joash Amupitan will be REJECTED!
INEC WILL SET NIGERIA ON FIRE!
#ArrestLereOlayinkaNow !!!
Pornography & Masturbation will hold you back from making meaningful progress in life…
Your best bet to speed in life is to stop watching pornography & masturbating.
Many are stagnated in life because of this;
Many lost favour in life because of this;
Many experience failure at the edge of breakthrough because of this.
You, Yes You, You know I am speaking to you…
It’s not worth it.
Missing a sick patient is far more costly than flagging a healthy one.
In the cause of this project I’ve learned firmly that domain knowledge is very significant to building an efficient machine learning model.
#DataScience#MachineLearning#Python#HealthcareAnalytics
BUILDING A “Heart failure Prediction Model” IN PUBLIC 3
Good morning #Datafam,
Following my last update on the actions and decisions I took regarding Data cleaning, Features Engineering and K-means clustering.
Prior to that,I took the following approach for Encoding and Scaling.
BUILDING A “Heart failure prediction model” in PUBLIC 2
Good day #Datafam,
This is where the machine learning model got more complicated and also the areas I learned that context and domain knowledge is very essential to machine learning project.
#python
https://t.co/ZLXysokpqb
The strongest overall approach was SMOTE + Threshold 0.3 on Logistic Regression.
SMOTE corrected the training imbalance so the model learned minority patterns better,and Threshold tuning then maximized recall at inference time,which in healthcare is the metric that matters most.
I am a British, Nigerian yes but I’ve always cared about Nigeria and anybody that has followed my career can vouch for that. I have houses in Nigeria, I want to take my kids home to a safe country just like we travel the rest of the world. Why should I not speak up about what is happening in my homeland?