Big Data Analytics

Applications in Business and Marketing

Kiran Chaudhary, Mansaf Alam

• Publisher: Auerbach Publications
• Year: 2021
• ISBN:
  9781000523577
• Edition: 1
• Topic: Business
• Subtopic: Marketing
• Index: Business

Big Data Analytics: Applications in Business and Marketing explores the concepts and applications related to marketing and business as well as future research directions. It also examines how this emerging field could be extended to performance management and decision-making. Investment in business and marketing analytics can create value through proper allocation of resources and resource orchestration process. The use of data analytics tools can be used to diagnose and improve performance.

The book is divided into five parts. The first part introduces data science, big data, and data analytics. The second part focuses on applications of business analytics including:

• Big data analytics and algorithm
• Market basket analysis
• Anticipating consumer purchase behavior
• Variation in shopping patterns
• Big data analytics for market intelligence

The third part looks at business intelligence and features an evaluation study of churn prediction models for business Intelligence. The fourth part of the book examines analytics for marketing decision-making and the roles of big data analytics for market intelligence and of consumer behavior. The book concludes with digital marketing, marketing by consumer analytics, web analytics for digital marketing, and smart retailing.

This book covers the concepts, applications and research trends of marketing and business analytics with the aim of helping organizations increase profitability by improving decision-making through data analytics.

Chapter 1 Embrace the Data Analytics Chase: A Journey from Basics to Business

Suzanee Malhotra

DOI: 10.1201/9781003175711-1

Contents

1.1 Overview

1.1.1 Data Science

1.1.2 Big Data

1.1.3 Data Science vs. Big Data

1.2 Data Analytics

1.2.1 Relationship Among Big Data, Data Science, and Data Analytics

1.2.2 Types of Data Analytics

1.2.2.1 Descriptive Analytics

1.2.2.2 Diagnostic Analytics

1.2.2.3 Predictive Analytics

1.2.2.4 Prescriptive Analytics

1.3 Business Data Analytics

1.3.1 Applications of Data Analytics in Business

1.4 Data Mining, Data Warehouse Management, and Data Visualization

1.4.1 Data Mining

1.4.2 Data Warehouse Management

1.4.3 Data Visualization

1.5 Insights in Action: Gains from Insights Generated out of Data Analytics

1.6 Machine Learning and Artificial Intelligence

1.7 Course of the Book

References

1.1 Overview

The coming age of business has introduced new terminologies in the business dictionary, some of which add ‘data science’, ‘big data’, ‘analytics’, and many more puzzling terms to the list. With the ‘data’ coming to the center stage of business, data collection, data storage, data processing, and data analytics have all become fields in themselves. Further, novel data keeps on adding to the previous data sets at humungous speeds. With rapid advances at the front of business, companies place data on the same pedestal as the other corporate assets, for it offers the potential and capabilities to derive many important findings. The sections following provide us with the meanings of data science and big data and a comparison of the two.

1.1.1 Data Science

With the data and data-related processes becoming more and more worthy, data science has become the need of the hour. Data science refers to scientific management of data and data-related processes, techniques, and skills used to derive viable information, findings and knowledge from the data belonging to various fields (Dhar 2013). It is a complex term that deals with collection, extraction, purification, manipulation, enumeration, tabulation, combination, examination, interpretation, simulation, visualization, and other such processes applied to data (Provost and Fawcett 2013). The various processes and techniques applied to data are derived from many different disciplines like computer science, mathematics, and statistical analysis (Dhar 2013). But it is not only limited to these disciplines and finds equal and substantial application in the fields of national defense and safety, medical science, architectonics, social science areas, and business management areas like marketing, production, finance, and even training and development (Provost and Fawcett 2013). In simple terms, data science is an all-encompassing term for tools and methods to derive insightful information from the data.

1.1.2 Big Data

Big data is often termed as “high volume, high variety and high velocity” data (McAfee and Brynjolfsson 2012). Big data is known as the enormous repository of data garnered by organizations from a variety of sources like smartphones and other multimedia devices, mobile applications, geological location tracking devices, remote sensing and radio-wave reading devices, wireless sensing devices, and other similar sources (Yin and Kaynak 2015). The global research and advisory firm Gartner considers “big data as high-volume, and high velocity or high-variety information assets that demand cost-effective, innovative forms of information processing that enable enhanced insight, decision making, and process automation” (Gartner Inc. 2021). Many organizations add another ‘v’, that is, veracity, to the definition of big data (Yin and Kaynak 2015). Big data represents the important and huge amount of data not amenable to traditional data-processing tools but with the potential to guide businesses to strategic decision-making from the important insights derived from it (Khan et al. 2017). Big data is categorized into structured, unstructured or semistructured types of data sets (McAfee and Brynjolfsson 2012). Structured data refers to well-organised and systematic data (like that once stored in DBMS software). The data that is simply stored in the raw version (like analogue data generated from a seismometer) without any systematic order or structure is known as unstructured data (Alam 2012b). In between these two lies semistructured data, where some part of data is unstructured and some structured (like data stored in XML or HTML formats).

Other types of data sets can be categorised on the basis of the time, viz., historical (or past information data) or current (novel and most-recently collected information data). On the basis of the source of data collection, data sets can be categorised as first-party data (collected by the company directly from their consumers), second-party data (purchased from another organization) and third-party data (the composite data obtained from a market square). Organizations often keep a customized and dedicated software for storage of big data, from which it can be easily put to computation and analysis to discover insightful trends from data in relation to various stakeholders.

1.1.3 Data Science vs. Big Data

With a basic understanding of these two data-revolutionizing ideas, let’s explain the boundaries separating these two.

Data science is an extended domain of knowledge, composed of various disciplines like computers, mathematics, and statistics. Contrastingly, big data is a varied pool of data from varied sources so huge in volume that it requires special treatment. Big data can be everything and anything, from content choices to ad inclinations, search results or browsing history, purchasing-pattern trends, and much more (Khan et al. 2015). Data science provides a number of ways to deal with big data and compress it into feasible sets for further analysis. Data science is a superset that provides for both theoretical and practical aid to data sorting, cleaning and churning out of the subset big data for the purpose of deriving useful insights from it. If big data is the big Pandora’s box waiting to be discovered, then data science is the tool in the hands of an organization to do such honours. Thus, one can say that, if data science is an area of study, then big data is the pool of data to be studied under that area of study.

After explaining these two upcoming concepts of both data science and big data, now let us turn our focus to the understanding of data analytics and its related concepts.

1.2 Data Analytics

Data analytics is the application of algorithmic techniques and methods or code language to big data or sets of it to derive useful and pertinent conclusions from it (Aalst 2016). Thus, when one uses the analytical part of data science on big data or raw data in order to derive meaningful insights and information, it is called data analytics. It has gained a lot of attention and practical application across industries for strategic decision-making, theory building, theory testing, and theory disproving. The thrust of data analytics is on the inferential conclusions that are arrived at after computation of analytical algorithms. Data analytics involves manipulation of big data to obtain contextual meanings through which business strategies can be formulated. Organizations use a blend of machine-learning algos, artificial intelligence, and other systems or tools for data-analytics tasks for insightful decision-making, creative strategy planning, serving consumers in the best manner, and improving performance to fire up their revenues by ensuring sustainable bottom lines.

1.2.1 Relationship Among Big Data, Data Science, and Data Analytics

Data, defined as a collection of facts and bits of information, is nothing novel to organizations, but its importance and relevance has acquired a novel pedestal in the current times. With global data generation growing at the speed of zetta and exa-bytes, it has indeed become an integral part of the business-management domain. Dealing with a mass of data existing in many folds of layers and cutting across many domains is the common link connecting data science, big data, and data analytics. Table 1.1 summarizes the interconnected relationship among big data, data science, and data analytics.

Table 1.1 Interconnected Relationship among Big Data, Data Science, and Data Analytics

Big Data →	Data Science →	Data Analytics
Big data is humungous in volume, value, and variated data gathered from different sources, requiring further dissection and polishing using data science and data analytics for important inferences to be derived from it.	Data science refers to a multidisciplinary field that involves collection, mining, manipulation, management, storage, and handling of the big data for smooth utilization and analysis of data.	Data analytics is an approach to derive trends and conclusions from the chunks of processed big data as made available after the initial mining and management processes run under the domain of data sciences for revealing intriguing and influential insights amenable to practical application.

1.2.2 Types of Data Analytics

It is vital to get a clear understanding of the different variants of data analytics available so as to leverage the stack of data for material benefits. The four variants of data analytics are descriptive, diagnostic, predictive, and prescriptive. The data analytics type is given in Figure 1.1. A combined usage of the different variants of data analytics and their corresponding tools and systems adds clarity to the puzzle—where the firm is standing and the journey to where it can reach by achieving its goals. A discussion regarding the four types is provided in the following paragraphs.

Figure 1.1 Types of Data Analytics.

1.2.2.1 Descriptive Analytics

As the name suggests, descriptive analysis describes the data in a manner that is orderly, logical, and consistent (Sun, Strang and Firmin 2017). It simply answers the question of ‘what the data shows’. It is further used by all the other types of data analytics to make sense of the complete data. Descriptive analytics collates data, performs number crunching on it, and present the results in visual reports. Serving as the primary layer of data analytics, it is most widely used across all fields from healthcare to marketing to banking or finance. The tools and methods applied in the process of descriptive analytics present the data in a summarized form. The data collated from a consumers’ mailing records, describing their mail ID, name, and contact details, is an example of it.

1.2.2.2 Diagnostic Analytics

As suggested by the name, diagnostic analytics looks into the reasons or causes of any event or happening and supplements the findings of the descriptive analytics (Aalst 2016). It simply answers the question ‘why or what led to any specific event?’ by delving into the facts to direct the future course of planning. It aims at first diagnosing the problems out of the data sets and then dissecti...