Question

1) Please give two different data formats and the suggested input types for data visualization. Please explain your reasons in detail regarding the effects, performance, and validation.

2) Please give an example for using a specific Google Chart. You need to explain in detail why that chart is a better choice than the other chart types.

Answer 1

1) DataFrame ( like .csv format) and List(with required indexing) data structure are most useful and widely used data types for better data visualization.

Effects of using DataFrame :

After processing any file in DataFrame format , while saving it as csv file it will include one more column as index which is very unnecessary. This data structure has only one side effect.

Performance of DataFrame :

In Data science and Machine Learning domain , DataFrame is working as very base and frequently used data structure. In ML and DS have too much priority on features of data. DataFrame has functionality to store data with feature. In DataFrame whole data store in series manner. On any particular feature on data , we can perform sorting , filtering and searching. This all the functionality increase the usefulness and performance of DataFrame data structure. You can visualize any data series by passing just name of feature if it is stored in DataFrame.

Validation of DataFrame :

There is no specific validation rules for fetching data and converting it into DataFrame. But while processing DataFrame , we must have to maintain or validate certain things like size of data and data type ( like string , int , float32 , float64 etc) . While fetching data from anywhere , in each columns may be there is more than one datatypes are available like ( int , float , char). It has also null values. we have to perform typecasting with iterators and filling some missing values is necessary to maintain integrity of DataFrame.(which is require for further visualization ).

Effects of using List :

Very high memory consumption with respect to array. To make certain operation fast like searching , sorting ,deletion and inserting a data , it needs to store many pointers like front side , rear side and many more.

Performance of List :

This data structure is very useful when we don't know how much data require to store while processing. It is dynamic data structure because while run time we can add data into it or delete it as per requirement. It has capability to add data into it in o(1) time. For Data visualization purpose we must have to set up appropriate index to plot points. Setting up index into List is very fast.

Validation of List :

We have to define structure of node( which is store the element ) . To maintain list operation fast each new added element have to follow node structure. Every list must store its head pointer address into variables and update it as per operation.

2) Example of using Google chart

Real time trading plotting , population analysis , plotting gold price variation with respect to time .

Why Google charts are better than other charts ?

Google charts are very customizable as compare to other tools like matlab , matplotlib or seaborn. Google charts very easy to integrate with web pages while developing analysis system , while in other charts are embedded only in certain type of web page. If client machine send continuously data into server and In reply server transform it into some plotting using google chart and sent it to the client machine. So google charts can respond better into live streaming. Google charts are free to use and you need just good internet conncetion while on other tools you have to install certain libreries and packages or software tools. Google charts can respond mostly every cloud.