Fundamentals of Data Science

Sanjeev J. Wagh, Manisha S. Bhende, Anuradha D. Thakare

• Publisher: Chapman and Hall/CRC
• Year: 2021
• eBook ISBN: 9780429811463
• Topic: Economics
• Subtopic: Data Mining
• Index: Economics

Fundamentals of Data Science is designed for students, academicians and practitioners with a complete walkthrough right from the foundational groundwork required to outlining all the concepts, techniques and tools required to understand Data Science.

Data Science is an umbrella term for the non-traditional techniques and technologies that are required to collect, aggregate, process, and gain insights from massive datasets. This book offers all the processes, methodologies, various steps like data acquisition, pre-process, mining, prediction, and visualization tools for extracting insights from vast amounts of data by the use of various scientific methods, algorithms, and processes

Readers will learn the steps necessary to create the application with SQl, NoSQL, Python, R, Matlab, Octave and Tablue.

This book provides a stepwise approach to building solutions to data science applications right from understanding the fundamentals, performing data analytics to writing source code. All the concepts are discussed in simple English to help the community to become Data Scientist without much pre-requisite knowledge.

Features :

• Simple strategies for developing statistical models that analyze data and detect patterns, trends, and relationships in data sets.
• Complete roadmap to Data Science approach with dedicatedsections which includes Fundamentals, Methodology and Tools.
• Focussed approach for learning and practice various Data Science Toolswith Sample code and examples for practice.
• Information is presented in an accessible way for students, researchers and academicians and professionals.

Part I

Introduction to Data Science

1

Importance of Data Science

DOI: 10.1201/9780429443237-2

1.1 Need for Data Science

Today lots of data are generated via various electronic gadgets. Earlier lots of data were stored in unstructured as well as structured format and lots of data flow happen. If you look today, data is created faster than it is imagined. While travelling by road, we can recognize lots of data being created, for example, vehicles speed, traffic light switching, Google map, etc. which get captured through satellites and transmitted to handheld devices in real time. It guides by showing several alternative paths with traffic intensity and a favorable path to take. We also get a suggestion on which route to take and which route to avoid. So there is a lot of data being created as we do our day-to-day activity. The problem is that we are not doing anything with the data. We are not able to analyze due to inefficient scientific insights.

We are creating data, but we are not utilizing it for behavioral analysis or predictions. If we look at the corporate world, we find that lots of data reports are generated. But what about utilization? How much time are we spending on preparation? We are not drawing any scientific insights; e.g., we are not using data for forecasting. It can be forecasted for sale in the coming months, seasonal time, or years. We can conclude from available data for decision making and prediction. From the data, we can visualize what the actual status of the data is and what decision we have to take. It is possible for shop owner to predict that if person wearing a red shirt and have age between 25 and 30, will be interested in the product. We can also see pattern discovery. There are lots of sales happening in November and December every year, that is, in the festival season, and also the changes in weather increase the sale, so actually we need to draw lots of pattern from data and do discovery. It needs to recognize the process of data analysis, data patterns and behaviors patterns using data science for increasing sale.

The upcoming automobile technology is an autonomous car. It is exciting to have a car driving automatically which will take you from home to office and office to home, and data science uses the coordinates to take a lot of decisions in this whole process. Decisions of whether to speed up, whether to apply the brake, whether to take left or right turn, and whether to slow down are a part of data science. The case study predicts that self-driving cars will minimize accidents, and in fact, it will save more than two million deaths caused by car accidents annually. Self-driving cars have a lot of scope for research design and testing in force to update knowledge of cars. Every automotive company is investing in self-driving cars. Studies say that in about 10–15 years, most of the cars will be autonomous self-driving cars [1]. The possible environment for autonomous vehicles is shown in Figure 1.1.

FIGURE 1.1
Autonomous vehicles.

Another example, while booking travel plan, passenger are unaware about the weather condition as per decided travel schedule, hence need data science for predicting and alerting passenger about weather condition for taking decision of travel. Using data science, proper route planning can be done if in advance weather condition is predicted. Also air traffic managements system can plan the alternate route by rescheduling the flights as per environmental conditions. If the route is not planned properly, then you might end up in a situation where the flight is not available. Planning alternate flights between cities are challenging in last minutes due to preschedule air traffics. The roles of data science in airline industries are shown in Figures 1.2 and 1.3. If we use data science properly, most of these problems can be avoided and it will help in reducing the pain both for the airlines and for the passengers.

FIGURE 1.2
Data science and airline industries.

FIGURE 1.3
Use of data science in an airline operating system.

Few more applications can be possible in the airline industry such as better route planning so that there will be fewer cancellations and avoid the frustration of pa...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Table of Contents
• Preface
• Author
• Part I Introduction to Data Science
• Part II Data Modeling and Analytics
• Part III Platforms for Data Science
• Index