Data Scientist

Given a string, write a program to count the number of vowels, consonants, and spaces in that string.

EXAMPLE :

Input: ‘N’= 25, ‘s’ =”Take u forward is Awesome”
Output: 10 11 4

Given a string say a = #raja#datascience#ml#credit,2#trivik#credit#ml#hacking#IEEE,3#abi#SEWA_dept#maths#resear,4#siva#IIT#fund_analytics#bi#portfolio#credit"  (dataset concatenated with '#')

Write a program to retrieve data from the string into a dataframe.

Why did you use the gradient boost algorithm for a crop yield production project?

You are given a string 'S'. Your task is to check whether the string is palindrome or not. For checking palindrome, consider alphabets and numbers only and ignore the symbols and whitespaces.

Note :

String 'S' is NOT case sensitive.

Example :

Let S = “c1 O$d@eeD o1c”.
If we ignore the special characters, whitespaces and convert all uppercase letters to lowercase, we get S = “c1odeedo1c”, which is a palindrome. Hence, the given string is also a palindrome.

Given a string 'S' and a list 'wordList' that consists of 'N' distinct words. Let 'Wi' denote word at index 'i' in 'wordList'. For each word 'Wi' in 'wordList', you need to determine whether it is present in string 'S' or not. Return a boolean array, where a boolean value at index ‘i’ represents whether the word ‘Wi’ is present in the string ‘S’ or not.

Given an input string 'S' and a pattern 'P', implement a regular expression matching with the support of two special characters ‘.’ (dot) and ‘*’(asterisk) where

1. ‘.’ matches to any single character.
2. ‘*’ matches to zero or more of the preceding element.

If the input string 'S' matches the pattern 'P', then return true else, return false.

Note:

1. You have to match the entire string with the pattern given.

2. Both the strings, 'S' and 'P' contain only lower-case alphabets.

3. Only the pattern will contain additional characters ‘*’ and ‘.’ along with alphabets.

Given a dataset of 1000 accounts that have applied for a loan, only very few, say 1% or 2%, could be filtered to get a loan.
What kind of model would you build, and which performance measures would you choose to determine the performance of your algorithm
For example, I have 100 accounts, only 2%, i.e., two accounts qualify to get a loan, and I build a model that says 100 accounts are disqualified for getting a loan. What is the accuracy of my model?

Could you tell me about your Machine Learning Package? (Learn)

Explain the KMeans algorithm in detail (which was included in my package). (Learn)