This article was published as a part of the Data Science Blogathon

## Overview

1. Learn the basic concept of Data mining
2. Understand the Applications of Data Mining

## Prerequisites

• Basic understanding of Python
• Basic knowledge of DataBase

Welcome guys!

Here I am going to give you a brief understanding of the basic concepts of Data Mining. We know that everywhere there is data in a various format which is to be stored in a database. According to the scale of data, we can choose a proper database. So there are popular databases that we know such as PostgreSQL, NoSQL, MongoDB, Microsoft SQL Server, and many more.

In this article, you will be getting an idea of Data Mining.

So Let’s move on…

## What is Data Mining:-

“Data Mining”,  that mines the data. In simple words, it is defined as finding hidden insights(information) from the database, extract patterns from the data.

There are different algorithms for different tasks. The function of these algorithms is to fit the model. These algorithms identify the characteristics of data. There are 2 types of models.

1)Predictive model

2)Descriptive model

Under this section, we are going to see some of the mining functions/tasks.

#### 1)Classification

This term comes under supervised learning. Classification algorithms require that the classes should be defined based on variables. Characteristics of data define which class belongs to. Pattern recognition is one of the types of classification problems in which input(pattern) is classified into different classes based on its similarity of defined classes.

#### 2)Prediction

In real life, we often see predicting future things/values/or else based on past data and present data. Prediction is also a type of classification task. According to the type of application, for example, predicting flood where dependant variables are the water level of the river, its humidity, raining scale, and so on are the attributes.

#### 3)Regression

Regression is a statistical technique that is used to determine the relationship between variables(x) and dependant variables(y). There are few types of regression as Linear, Logistic, etc. Linear Regression is used in continuous values(0,1,1.5,….so on) and Logistic Regression is used where there is the possibility of only two events such as pass/fail, true/false, yes/no, etc.

#### 4) Time Series Analysis

In time series analysis, a variable changes its value according to time. It means analysis goes under the identifying patterns of data over a period of time. It can be seasonal variation, irregular variation, secular trend, and cyclical fluctuation. For example, annual rainfall, stock market price, etc.

#### 5) Clustering

Clustering is the same as classification i.e it groups the data. Clustering comes under unsupervised machine learning. It is a process of partitioning the data into groups based on similar kinds of data.

#### 6) Summarization

Summarization is nothing but characterization or generalization. It retrieves meaningful information from data. It also gives a summary of numeric variables such as mean, mode, median, etc.

#### 7) Association Rules

It’s the main task of Data Mining. It helps in finding appropriate patterns and meaningful insights from the database. Association Rule is a model which extracts types of data associations. For example, Market Basket Analysis where association rules are applied to the database to know that which items are purchased together by the customer.

#### 8) Sequence Discovery

It is also called sequential analysis. It is used to discover or find the sequential pattern in data.

Sequential Pattern means the pattern which is purely based on a sequence of time. These patterns are similar to found association rules in database or events are related but its relationship is based only on “Time”.

Up to this point, we have seen all the basic functions or tasks of Data Mining. Let’s go-ahead to know more about Data Mining…

## Data Mining VS KDD(Knowledge Discovery in Database)

Data Mining: Process of use of algorithms to extract meaningful information and patterns derived from the KDD process. It is a step involved in KDD.

KDD: It is a significant process of identifying meaningful information and patterns in Data. The input is given to this process is data and output gives useful information from data.

KDD process consists 5 steps:

1)Selection: Need to obtain data from various data sources, databases.

2)Preprocessing: This process of cleaning data in terms of any incorrect data, missing values, erroneous data.

3)Transformation: Data from various sources must be converted, encoded into some format for preprocessing.

4)Data Mining: In this process, algorithms are applied to transformed data to achieve desired output/results.

5)Interpretation/evaluation: Has to perform some visualizations to present data mining results which are very important.

## Data Mining Applications

#### 1) E-Commerce

E-commerce is one of the real-life applications of it. E-commerce companies are like Amazon, Flipkart, Myntra, etc. They use data mining techniques to see the functionality of every product in such a way that “which product is viewed most by the customer also what they also liked other”.

#### 2) Retailing

It is another application of data mining from the retail market. Retailers find the pattern of “Freshness, Frequency, Monetary(In terms of Currency)”.  Retailers keep the track of sales of products, transactions.

#### 3) Education

Education is an emerging, trending field nowadays. It concerns knowledge discovery from educational data. The main goal of this application is to study or identify the student’s behavior pattern in terms of future learning, effects of study, advanced knowledge of learning, etc. These data mining techniques are used by institutions to take accurate decisions and also predict appropriate results.

– KNIME

-WEKA

-ORANGE

### Data Mining Algorithms

•  K-means clustering
• Support vector machines
• Apriori
•  KNN
• Naive Bayes
• CART and many more…

These are few algorithms.

*Now I am going to give you information about the required libraries below.

– Apriori:

`from apyori import apriori`

– K-means clustering:

``````from kneed import KneeLocator
from sklearn.datasets import make_blobs
from sklearn.cluster import KMeans
from sklearn.metrics import silhouette_score
from sklearn.preprocessing import StandardScaler``````

– Support Vector Machines:

`from sklearn import svm`

-Naive Bayes:

`from sklearn.naive_bayes import GaussianNB`

-CART:

`from sklearn.tree import DecisionTreeRegressor`

-KNN:

``from` `sklearn.neighbors ``import` `KNeighborsClassifier``

So here are few libraries that to be installed while performing the algorithm.

### Conclusion

I hope you liked my article. If you have any queries, you can drop comments below. Thank You!

The media shown in this article are not owned by Analytics Vidhya and is used at the Author’s discretion. 