Introduction to Data Science

• Data Science involves the collection, analysis, and decision-making related to data.
• The goal of data science is to identify patterns in data through analysis and predict future events.

Employing Data Science

• Businesses can make better decisions.
• Predictive analysis can be used to identify what is going to occur next.
• Data science helps find patterns in data and identify hidden information.

Importance of Data Science

• Analysis of data requires competent practitioners to provide actionable insights.
• Many industries use data science, including banking, consultancy, healthcare, and manufacturing.
• Demand for data scientists and data engineers has tripled, rising 231% in five years.

The 3V Model for Data

• Velocity refers to the speed at which data is accumulated.
• Volume is the size and scope of the data.
• Variety is the array of data and types (structured and unstructured).
• The increase in size of data is a 50-fold volume increase from 2010.

Acritical Intelligence

• Includes Vision, Voice recognition, Natural language processing, Robotics, Neural Network.

Computer Vision

• Branch of AI which uses digital images, videos, and other visual inputs to allow computers and systems to extract useful information.

Voice Recognition

• Voice recognition has become more popular and useful using AI.
• Ex: Amazon's Alexa and Apples Siri.

Natural Language Processing

• Helps computers in understanding how people write and speak,

Robotics

• Robotics' goal is to build intelligent robots using AI.

Neural Network

• A technique in Artificial Intelligence that trains machines to process data inspired by the human brain

Data Science vs Acritical Intelligence

• Data science uses AI because of the volume of data that can't be handled; machine learning is utilized.
• Data Science Output can be analysis (past or present) or analytics (predicted).

Analysis

• In the dataset, the values are sorted.
• This dataset contains the sorted list of heights and weights of people
• Heights range from 58 to 72, weights range from 115 to 164
• Data indicates weight increases with height.

Analytics

• Machine learning algorithms can be used to predict the weight of a person given their height.

Skills for Data Scientists

• A strong knowledge of basic statistics and AI is needed.
• Computer science skills are needed to handle complex datasets with programming languages like R or Python.
• The ability to visualize and express data and analysis in a meaningful way is needed.

Ethics, Bias, and Privacy in Data Science

• Issues can be traced back to the origin of the data, requiring considerations of collection methods and intended use.

Requirements to be a Data Scientist

• Data Skill: Database, SQL and Hadoop or Spark
• Programming Skills: Python, R
• Other Requirements: Cloud computing, Data pre-processing, Data visualization, Deep learning, Machine Learning, Model deployment

Typical Data Science Process

• Data Gathering collects and analyzes data from multiple sources to find insights.
• Data Analysis extracts meaningful information from data for making conclusions.
• Data Preprocessing involves cleaning, converting, and combining data to prepare it for analysis.
• Predictive Analytics predicts future events using data analysis, machine learning, AI, and statistics.
• Knowledge Extraction extracts knowledge through machine learning, natural language processing, and data mining.
• Data Visualization uses standard images such as charts, plots, infographics, and even animations.
• Business Application helps businesses in obtaining comprehensive market, competitive, and consumer information.

Machine Learning

• An AI application that uses statistical methods for computers to learn and make decisions without being programmed.
• Used when: Human expertise doesn't exist, humans can't explain expertise, solutions change in time, solutions need to be adapted.
• Consists of Supervised learning and Unsupervised Learning

Supervised Learning

• Supervised learning uses well-labeled training data to predict outputs.
• Two types: classification and regression.

Classification

• Used when output is categorical with two or more classes.

Regression

• Regression is used when the output variable is a real or continuous value.

Unsupervised Learning

• Machine learns from unlabeled data to discover patterns known as Clustering.

Data Scientists vs Machine Learning Engineers

• A Data Scientist has Machine Learning knowledge to some extent but Machine Learning Engineers have it as domain knowledge.
• Data scientists should have extensive domain knowledge in the fields in which they work.
• Machine Learning Engineers work under the vision of Data Scientists.