Machine Learning Process

27 / 38

Please select all the columns which you think can be used as features.

Remember, our task is to build a machine learning model to predict what sorts of people were likely to survive.


Titanic Dataset

See Answer

Note - Having trouble with the assessment engine? Follow the steps listed here


No hints are availble for this assesment

Please login to comment

30 Comments

Sorry, I am unable to understand why 'Name' is not an attribute. 

  Upvote    Share

Good question! The reason "Name" is not considered a useful attribute (feature) in most machine learning models is due to it's Lack of Predictive Power. A person’s name does not have a direct relationship with survival probability. 

  Upvote    Share

Thanks, Shubh.

 

However I want to report a bug in your online application. 

I have already completed this section  ,till the end and marked it as completed . when I  logiged in  today then application has took me to question number 3 . Is it that this section is not recorded as Completed at backend. 

  Upvote    Share

Hi Kulwinder,

Thank you for bringing this to our notice. We will surely look into it.

  Upvote    Share

how is parch a feature? it has the same value across all instances

  Upvote    Share

Hi Harshit,

In above case, we are just showing 5 records from the whole dataset. So, in other records, patch may contain multiple values. Parch is the number of Parents/Children Aboard. You can see the whole dataset at: https://www.kaggle.com/competitions/titanic/data

 1  Upvote    Share

Except name and Passenger id

 2  Upvote    Share

Hi,

Please select the respective options above and then click on the Submit Answer button to complete this assignment.

Thanks.

  Upvote    Share

How does embarked matter? 

  Upvote    Share

Hi,

Embarked denotes the port of embarkation, here C = Cherbourg, Q = Queenstown, S = Southampton. I believe the allocation of berths depends the same. You can find more about the features from the Kaggle page of this dataset:

https://www.kaggle.com/c/titanic/data

Thanks.

  Upvote    Share

I am not getting the answer, can anybody tell me correct answer.

  Upvote    Share

Hi,

The answer is available below the question.

Thanks.

  Upvote    Share

hello there is no option to see the result ?? why is that so?

 

  Upvote    Share

Hi,

This question is to test the learner's understanding. Please go through the problem statement and answer the question.

All the best!

  Upvote    Share

why not Parch ?

  Upvote    Share

Hi,

Parch is the number of Parents/Children Aboard. Does it contribute towards survival rate? What do you think?

Thanks.

  Upvote    Share

Then please remove Parch from the answer list.

Kindly correct the list of answers, I wanted to know why Parch is among the answers.

"Age, Sex, Embarked, Pclass, SibSp, Parch, Fare" on selecting this I was able to proceed further. Try submitting these answers.

Kindly check. Thanks

  Upvote    Share

Thanks, I always follow your responses. Really usefull and equally important for learning.

  Upvote    Share

Hi,

Thank you for your kind words. Parch is among the answer because it is one of the features that contribute towards the survival rate.

Thanks.

  Upvote    Share

i have all tried all answers but still showing wrong 

  Upvote    Share

This comment has been removed.

Of all the available options, I selected : Age, Name, Sex, Embarked, Pclass, Sibsp, Fare (PassengerID - not selected as it is unique for each instance. Parch - not selected since all instance have same value of 0). Still it says Incorrect answer. Can you please correct me where am I going wrong?

  Upvote    Share

Hi,

Can you use Name as a feature? How will you use names to predict if that person can survive or not? These questions should help you choose the correct options.

Thanks.

  Upvote    Share

Hi Team,

As per the dataset, other than "survived" all of the coloumns can be used as features but even if I select all or a set of features such as 

1. Age,Sex,Embarked,SibSp,Parch

2. Age,Sex,Embarked,SibSp,Parch,Fare

It does not point me to the answer. Please help out here on the actual answer.

  Upvote    Share

Hi,

The question asks out of all the columns listed above, which are the ones you can use as features. Can you use all of these columns as features?

Thanks.

  Upvote    Share

hi,
in this question, for me looks like Age, Sex, PClass,SibSp and Parch should be sufficient features for modelling. Please help in better understanding.

  Upvote    Share

Hi,

The question here has asked to list all the features that can be used for modelling, not the minimum number of features.

Thanks.

-- Rajtilak Bhattacharjee

  Upvote    Share

Yeah, but in order to solve the problem and around the context, why do we need fare. If we include Fare then we can have Name as well. I am trying to understand the significance within the context of the given problem. Appreciate clarifications.

  Upvote    Share

Hi,

The Fare has a direct correlation with Class. And even though we are not considering Name here, it too can be used as a feature. How? The name, if you notice it, contains Mr. Miss. etc to denote gender/status. So we can easily infer from the data whether more women or men survived this tragedy. However, for that we need to extract the feature which is beyond the scope of this topic, and also it might not be much of an useful feature since not all passengers have used same salutation. So we have not included Name even though we have included Fare.

Thanks.

-- Rajtilak Bhattacharjee

  Upvote    Share