Data science-Deep Dive

Descriptive Analytics:

Descriptive analytics is the analysis of past (or historical) data to understand trends and evaluate metrics over time. This is the easiest technique of data analysis because it requires minimal to no coding at all. There are many sophisticated tools already existing that can handle descriptive analytics.

Predictive Analytics:

Predictive analytics predicts future trends. It is crucial to remember that predictive analysis only forecasts the future and not actually predict it with a hundred percent accuracy. It can also include predicting the values of missing fields in a data set and probable impact of data changes on future trends. Sentiment analysis and credit score are examples of predictive analytics.

Prescriptive analytics:

Prescriptive analytics showcases variable solutions to a problem and the impact of considering a solution to future trends. Prescriptive analytics is an evolving technique and there are limited applications for it in business. A self-driving car is a perfect example of prescriptive analytics.

Descriptive vs predictive vs prescriptive:


Descriptive predictive prescriptive
Tools used Data mining and data aggregation Simulation and statistical model optimization model and Heuristics
Limitation Limited ability to guide decisions Help inform low complexity decision Most effective where you have more control over what is being modeled
When to use Summarize result for all or part of your business When you want to make an educated guess at likely results when you have important, complex or time-sensitive decisions to make


Difference between Analytics vs. Analysis

Data analysis is a broader term that refers to the process of compiling and analyzing data in order to present findings to help the management in decision making. Data analytics is a subcomponent of data analysis that involves the use of technical tools and data analysis techniques.




Data analytics is a general form of analytics which is used in businesses to make a decision from data which are data driven Data analysis is a specialized form of data analytics used in businesses to analyze data and take some insight into it.  


Data analytics consists of data collection and inspect in general and it has one or more users. Data analysis consists of defining data, investigation, cleaning and transforming the data to give meaningful outcome.  


R,python,Tableau public,SAS,Excel Rapidminer,Tableau public,openrefine,NodeXL,google fusion tables are used.  


To find masked patterns, anonymous correlations, customer preference, and market trends one can perform analysis like exploratory analysis, descriptive analysis, predictive analysis and take useful patterns from the data.  


Let say you have 1gb customer purchase related data of past one year, one has to find that next possible purchase, you will use data analytics for that. let say you have the 1gb customer purchase related data of past one year and one should try to find what happened so far that means data analysis  


Supervised vs unsupervised learning

Having a full set of labeled data while training an algorithm is called supervised learning. In supervised learning, a neuron is provided with a dataset consisting of input vector and target associated with each input vector.

The aim of unsupervised learning is to discover features in the input data with no assistance from the external source. ( training will be provided to the machine).


Supervised learning

Unsupervised learning

Input data

Uses known and labeled data as input Uses unknown data as input

Number of classes

Number of classes are known

Number of classes are unknown

Computational complexity




Use offline analysis

Use Real-time analysis of data

Accuracy of result

Highly accurate

Less accurate


Data mining and it’s techniques

Data mining is the process of extracting knowledge or pattern from a large amount of data. Data mining is also termed as Knowledge Discovery in Database(KDD).

The knowledge Discovery process involves Data cleaning, Data integration, Data selection, Data transformation, Data mining, Pattern evaluation, and Knowledge presentation.


Tracking pattern -One of the most basic techniques in data mining is learning to recognize the pattern in your data.

Classification– It is a more complex data mining technique that forces you to collect various attributes together into categories.

Association– It is more specific to dependently link variables.In this case, you will look for specific events that are correlated with another event or attribute.

Clustering– Similar to classification but involves grouping of data together based on their similarities

Outlier detection– It is a process of simply recognizing the overarching pattern which cant give you a clear understanding of your data.

Regression– It is used to identify the likelihood of certain variable with the presence of other variables.

Prediction– It is the most valuable technique. Just recognizing and understanding the historical trends to chart a somewhat accurate prediction of what will happen in the future.