Better Alogirthms to predict who will survive?

[barcelonacode]

Need tuning of algorithms and better algorithms needed to impore accuracy.

[rahul21198]

@barcelonacode we can use Decision tree and Neural network to get better accuracy

[rahul21198]

I would like make these improvement s in this code :

Better data cleaning.

Droping the cabin data making use of progressive technique to find out the mean from the given data and replacing the unknown values add new graphs providing better understanding : sex-count Pclass-sex Pclass-person(male,female,child)using age Count-age group Adding Kernel density estimation plot to for male/female age density cabin-count Pclass-survived Chances of survival under different circumstances. Analysing the factors using decision tree. Using one hot encoding instead of number for better analytical result

[barcelonacode]

How will you replace Gender with numbers? ref line No: [163]

[rahul21198]

i will be replacing male and female using one hot encoding technique. male->[1,0,0] female->[0,1,0]

[barcelonacode]

Why are you using one hot encoding

[rahul21198]

REFER - https://machinelearningmastery.com/why-one-hot-encode-data-in-machine-learning/ for better understanding of it .

[rahul21198]

Moreover better datacleaning has lead to more distinct defined results .

[barcelonacode]

This might work better.