What you need to start learning data science

Before I went into data science, I thought that I should gain some knowledge in mathematics, so I planned to concentrate on core concepts in statistics, probabilities, calculus, and algebra. Without doubtful those concepts plays a major role in data science.

ANALYSIS VS ANALYTICS

Analysis(Past Events): Things that have already happened like explaining how a business problem has been solved or how a company faced financial losses.

Analytics(Future Events): is responsible for possible future events. it covers two types they are,

Qualitative Analytics: based upon instincts and analysis like suggesting a product to the customers.

Quantitative Analytics: Generally helps to predict future events, applying formulas and algorithm to our analysis. Example: Based upon the past existing data and knowing when to launch our product like selling "Ac in summer" and selling "Sweaters in Winters".

Qualitative Analytics and Quantitative Analytics can also be applied in Analysis.

Business Analytics:

business case studies: A company that examines its profit and loss in the past few months. So this activity has been done in business analysis(past) phases.

Business Intelligence (Bi):

Bi is the process of analyzing and reporting historical business data. So It deals with past data.

business case studies: A company that examines its profit and loss in the past few months. So this activity has been done in business analysis(past) phases.

Business Analytics:

1. Business Intelligence (Bi):

Bi is the process of analysing and reporting historical business data. It deals with past data.

2. Machine Learning(ml):

Machine learning is used to create an algorithm.

No Data = No Decision.

Ml can be implemented with data. Ml holds data from 3rd party (Facebook) and makes a new pattern to provide a recommendation. It is used for encryption like "voice recognition, image recognition". In other words, ml has the ability of machines to predict future outcomes. In general, AI decides with computers and stimulate human knowledge. Machine learning is applied to Artificial Intelligence for human Ml helps to reach AI.

role in Data Science:

In a company, they provide a huge amount of data set from customer data is given. we use some analytics tools and extract useful knowledge to increase sales growth.

Data Science has five stages:

1. Traditional Data.[Data Collection]

2. Big Data.[Data Collection]

3. Business Intelligence.

4. Data science Technics.

5. Machine Learning Technics.

1. Traditional Data:

• Traditional Data represents the data in the form of "Numeric(Integer) or Text Values".

• Format: Structured and stored in Database.

Traditional Data - Data Preprocessing:

Data preprocessing helps us to evaluate Incorrect or Invalid data can be corrected. Data preprocessing Technics involves Class Labelling and Data Cleansing

i. Class labelling:

Arranging data by category like 'numerical' and 'categorical data'. numerical data contains numbers so it can be manipulated. categorical data contains words so it cannot be manipulated.

ii. Data cleansing:

• Data cleansing is also said to be data cleaning or data scrubbing. It deals with 'inconsistent data' like misspelt words (Norht India, Suoth India).

• Data cleaning helps to handle missing values.

Working with traditional data :

Data Shuffling :

Data shuffling was like shuffling a deck of cards used to prevent unwanted patterns, it improves predictive performance, this process can be done by shuffling all the data and taking a group of random data.

2. Big Data : [data collection]

Data preprocessing:

Class labelling collects such data as a number, text data, digital image data, digital video data, digital audio data.

Example: Social media tracks user's personal data, photos, and videos. So the volumes(V) of data stored are Extremely large.

Data masking is the process of storing confidential or personal information in a secured place. In other words "hiding original content with modified content".

3. Business Intelligence Analysis :

• Observation : Each value(user) is taken as one observation.

• Measures: Sum that is related to past data performance.

• Metrics: metrics are a value that is obtained from measures and metrics are useful for 'comparisons'.

BI - Key Performance Indicators(KPI) :

In KPI all possible metrics are extracted from the data set.

Key = business goals.

Performance = task performed in a specified duration.

Indicators = values or metrics related to our business.

• Example: metrics - who have visited our site; KPI - who have clicked our ad.

Thank you ...