The mammoth increase in data volumes had led to an information knowledge revolution triggered. Key aspect of research and strategy formulation to gather meaningful information and insights into the data already available is primary focus. All this information is gathered and stored in a data warehouse which is later used for business intelligence purposes. Data mining and Data analytics are two subsets of business intelligence.

Today we look more in detail about two most popular and widely used terminologies data mining and data analytics and how they differ from each other , purpose for which they are collected and their use cases.

About Data Mining

Since the 1990s data mining has been a buzz word which is a systematic and sequential process for identification and discovery of hidden patterns and information in large data sets. Data mining helps to transform raw data into useful information which could help businesses to achieve their goals. Data mining tasks require various techniques and tools such as data cleaning , Artificial intelligence, association rule, clustering, classification etc. Let’s look at each one of them more in detail.

Data cleaning – all raw data is taken from various sources and converted into a specific standard format which can be processed and analysed.
Artificial intelligence – system performs analytical activities such as planning, learning and problem solving
Association rule – is also called market basket analysis. And it is used to determine the relationship between different data set variables.
Clustering – refers to the process of splitting huge data sets into smaller and manageable subsets called clusters.
Classification – is used to assign categories to data collection to obtain more in-depth analysis and prediction

About Data Analytics

Data analytics is a superset of data mining which involves extraction, cleaning, transformation, modelling and visualization of data in a meaningful and useful way. It helps in deriving a conclusion and making informed decisions. Data analytics has been a process since the 1960s.

Data analytics goes through various phases as under :

Discovery – Analyse the data and investigate the problem related to data to develop a context and understanding.
Data preparation – An analytics sandbox is made to accomplish the analysis’s process for project duration. Various tasks are carried out such as extraction, transformation and data update in the sandbox.
Model planning – process and techniques are determined for a specific model building phase. Data sets are analysed to learn about relationships between various variables and choose most suited model
Model building – data sets are created for different purposes such as test, production, training etc
Communicate results – interaction with stakeholders and find if output of project is failure or success
Operationalize – outcome is delivered

Comparison Table: Data Mining vs Data Analytics

Below table summarizes the differences between the two:

FUNCTION	DATA MINING	DATA ANALYTICS
Definition	Process of extraction of specific pattern from larger data sets	Process of ordering and organization of raw data in such a manner that it provides useful insights and help in making informed decisions
Expertise area	Involves intersection of machine learning , statistics and databases	It requires knowledge of computer science, statistics, mathematics, subject knowledge and AI/Machine learning capabilities
Synonyms	Also known as knowledge discovery in databases (KDD)	Data analysis could be explanatory, descriptive, text analytics, predictive analysis, data mining etc.
Work Profile	Data mining professionals build algorithms to gather meaningful structure in data.	A data analyst may not be a single individual. The profile of data analyst involves multiple tasks such as raw data preparation, cleansing, transformation and modelling , presentation in pictorial format or non-chart-based visualizations
Responsibilities	Responsible for extraction and discovery of meaningful patterns and structure in data	Responsible for development of models, explanations, tests and analytical measures to propose hypothesis
Output	Data pattern is output from mining, trends	Verified hypothesis
Output	Data pattern is output from mining, trends	Actionable insights into data
Datasets	Large (Big data)	Small, medium, large
Datasets	Structured	Structured, unstructured, semi-structured
Goal	Make data usable	Aid in data driven decisions
Goal	Identification of pattern	Conclusions
Examples	e-commerce sector	Time series study of unemployment during last 15 years