January 8, 2021

Decision Making with Artificial Intelligence

Authored by Sri Divya Muktha K & Team

Explore what goes behind the scenes of AI and Data Science


The era of collecting data, storing, and analyzing through spreadsheets has gone far behind. Today huge and diversified amounts of data are collected through various social media like Twitter, Facebook, and mainly through online shopping sites. Collection of this type of data and analyzing it is not possible through traditional methods like Data Mining, SQL Querying, etc. Hence there is a need for the application of new algorithms to make sense of this increasingly huge quantity of data flowing in real-time. The ability to perform real-time analysis of data for providing insights on trends in market, demand, and supply chain and the targeted audience is essential. Data Science and Artificial Intelligence are buzz words of this generation that are accepted or will be adopted soon by most of the enterprises. Artificial Intelligence is a domain of study which makes machines intelligent and enables decision-making based on the data provided, while Data Science helps to generate meaningful insights from data.

Data Science

Extracting evocative insights from diversified data is the study of Data Science. This arena involves a wide variety of methodologies, technologies, and algorithms. It demands scalable hardware, technical skills, and efficient algorithms to find solutions to problems through data analysis. The evolution and revolution of technology to move towards data science is depicted in Figure 1.

Figure 1 . Evolution and Revolution of Data Science
Figure 1 . Evolution and Revolution of Data Science

This emerging methodology of Data Science is based on seven different elements. These include

  1. Statistics: This is the essential element of Data Science. This domain of knowledge is the building block of any analytics. Statistics is a mathematical science related to data collection, analysis, interpretation, and presentation of outcome. Analytics done using this can be qualitative or quantitative in nature. Identifying patterns or trends using graphs is the methodology of data interpretation in Quantitative Analytics. Generic knowledge extraction using visual, verbal and text data is the study as part of Qualitative analytics.
  2. Experts with Domain Knowledge: Data science is a branch of the subject that can be applied to any domain for doing various activities like prediction, visualization, analyzation, interpretation, etc. When it has to be applied to a specific domain, the subject matter experts in that domain are the key element in the study.
  3. Engineering on Data: Data engineering is an important and fundamental dexterity in any data science toolkit. The reason for this is that only the engineering domain will help or make people evaluate their skills and check whether the skills are aligned with the needs of the organization. This element helps data scientists to take on larger and more ambitious projects that are otherwise not reachable.
  4. Visualization: In this element, data is represented using graphical tools like charts, maps, etc., This element helps in understanding underlying patterns, trends and out of the box data. Visualization plays an important role in decision making. This is becoming an important key tool to make sense of trillions of rows of data generated every day. Results obtained through visualization of existing data will help organizations to create new hypotheses and even help in realizing the next step of action in the current scenario.
  5. Advancement in Computation: This element helps in creating a system that combines hardware and software for making computers to operate, further helping with designing, debugging, creating code for a specific task.
  6. Mathematics: This element plays an important role while dealing with data science applications. Important topics of mathematics that play a major role are linear algebra, differential calculus, discrete maths, and numerical analysis. Smart utilization of these techniques and applying it to the problems of data science is a key point of focus in this element.
  7. Machine Learning: This element helps in designing automated analytical models for data analysis. Focuses on making computers act without explicitly programmed. It helps in the creation of self-driving mechanisms in any device. Learn by Experience is the motto of functionality created with the machine learning element of data science. This becomes a basic component of Artificial Intelligence.

Figure 2 helps to understand about the amalgamation of all these elements of data science.

 Figure 2. Amalgamation of Elements into Data Science
Figure 2. Amalgamation of Elements into Data Science

Key skills to be a Data Scientist:

Knowledge and skills for working in Data Science domain needs the following tools

  • Data Analysis tools include R, Python, SAS, Matlab RapidMiner, etc.,
  • Data Storing tools include SQL, Hadoop, AWS Redshift, etc.,
  • Visualization tools include Cognos, R, Tableau
  • Machine Learning tools included are Spark, Azure, etc.,

Data Science Applications Lifecycle & Examples

Like any other applications or domains, Data Science design also must follow the following phases of its life cycle which is depicted in Figure 3.

Figure 3. Life Cycle of Data Science Domain Applications
Figure 3. Life Cycle of Data Science Domain Applications

Data Science is a multi-faceted domain of study. At present Data Science is used in applications like

  • Recognition of Text, Voice, and Image
  • Gaming
  • Surfing and Searching Internet
  • Prediction in Healthcare
  • Recommendation
  • Identification of Risks in Business

Data Science Limitations & Challenges:

Challenges while working with Data Science include

  • Finding and Identifying correct data with the correct size at the right moment of time
  • Consolidation and Association of Data to Create Information
  • Making people believe in Data
  • Creating Apt Models to Solve the Problems
  • Finding the right Case Studies for Analytics
  • The agility of the Model and Data involved in Model
  • Providing Security to the Data Retrieved

Artificial Intelligence

The preparation of intelligent machines is the study of artificial Intelligence. Intelligent machines mean machines that behave similarly to humans, react like humans, learn like humans, give reasoning like humans, and finally solve problems like humans. All this speaks about a “Machine with human skills and power”.  Learning AI means knowing what is intelligence and how it is composed, so as to do all brain-related activities together at the same point in time. Making machines behave and work like humans need knowledge of the following domains and disciplines of study which is shown using Figure 4. Domains include

  • Mathematics
  • Psychology
  • Neural Study
  • Statistics
  • Biology

Figure 4. Areas for Building AI applications
Figure 4. Areas for Building AI applications

Applications of Artificial Intelligence:

AI is an interdisciplinary study to create a machine with human power. Creation of such a machine helps in

  • Doing work with High Accuracy
  • Doing work with Minimal Errors
  • Performing with High Speed
  • Working in place of Human Risk
  • Assistance through Digitalization
  • High Reliability

Types of Artificial Intelligence:

Major contributions of Artificial Intelligence to the world include the Replication of humans, solving problems using knowledge intensively, connecting to various sources, and performing actions cleverly. This system contains various agents and environments in which these agents work. The Agent of Artificial Intelligent system works to observe its environment with the help of sensors and acts through effectors. These agents can be classified as Human-agent, Robotic agents and software agents.

Artificial Intelligence can be broadly classified into three levels

  1. Narrow Artificial Intelligence [4]
  2. General Artificial Intelligence [4]
  3. Strong Artificial Intelligence [4]

At present research and work of Artificial Intelligence are at level 1 called narrow Artificial Intelligence. The subfields of Artificial Intelligence include Machine learning and deep learning which is depicted in Figure 5.

Figure 5. Sub domains of Artificial Intelligence
Figure 5. Sub domains of Artificial Intelligence

In Practice Artificial Intelligence can be applied in the following applications like

  • Self-Driving Cars
  • Speech Recognition Applications using Alexa or Siri
  • Artificial Intelligence in games like Cricket, Chess, etc.,
  • Home Automation System
  • Robotic Applications using IBM Watson

Artificial Intelligence and Data Science

Artificial Intelligence will help in understanding data and helps in identifying relationships in data. The ability of Artificial Intelligence systems to learn by fetching patterns from existing data is called Machine Learning. Machine learning is a subset of Artificial Intelligence which is based on a concept that says that machines will help in accessing data and even have the capability to analyze and learn from data. Data Science and Machine learning work together for the same cause. Data Science helps to estimate the data for Machine Learning.  As said above Machine Learning is a subset of Artificial Intelligence. Data Science is an interdisciplinary domain to fetch knowledge from data. Figure 6 depicts the convergence of these three domains of study.

Figure 6. Convergence of Artificial Intelligence, Machine Learning and Data Science
Figure 6. Convergence of Artificial Intelligence, Machine Learning and Data Science

Using Figure 6. We can say that Artificial Intelligence is a discipline that can be combined with Data Science and new applications can be created. For example in the domain of Agriculture, Artificial Intelligence and Data Science can together assist the farmers with crop disease detection and pest management. Traditionally farmers get into losses due to invasion of pests like locust etc, To counter the damages due to these pests, farmers use pesticides. This has to be done at the right time, otherwise, the efforts go in vain, while overuse of pesticides can be counterproductive. This is where Artificial Intelligence and Data Science can come together to provide a solution. While Data Science can be used to define when and how much pesticide to be used, usage of drones can be employed to take images of the crop to correlate with the change in color of the leaves. Based on the change in color, one can diagnose the state of damage to the crop, and recommended solutions can be generated using AI.


Data is like food for today’s and tomorrow’s world. With the use of the right tools, technologies, and algorithms the raw data can be converted to meaningful and distinctive business development. These new upcoming domains help in building better and fastest decision-making systems. Data Science domain helps the Artificial Intelligence domain to find appropriate and meaningful information from the input data in an efficient manner. Taking this meaningful data from the Data Science, the Artificial Intelligence domain makes decisions in real-time. These two domains of the study help to automate many applications, by providing accurate solutions at the point of need. Scientists who work with data are like backbones to most big companies which are by nature data-intensive. In the near future, engineers and technologists who can work efficiently with the Data Science, Machine Learning, and Artificial Intelligence domain will notice numerous demanding opportunities. The actual challenge for these domains is Machine Learning - where we have petabytes of data generated every single day, it is essential to capture the right data while building data models using data science. These data models act as pillars of AI implementation in any field.


  1. (n.d.). Retrieved January 08, 2021, from https://www.cs.bham.ac.uk/research/projects/poplog/computers-and-thought/chap1/node8.html
  2. Chang, R. (2018, June 24). A Beginner's Guide to Data Engineering - Part I. Retrieved January 08, 2021, from https://medium.com/@rchang/a-beginners-guide-to-data-engineering-part-i-4227c5c457d7
  3. Data visualization beginner's guide: A definition, examples, and learning resources. (n.d.). Retrieved January 08, 2021, from https://www.tableau.com/learn/articles/data-visualization
  4. Essential Math And Statistics For Data Science Tutorial. (2020, May 14). Retrieved January 08, 2021, from https://www.edureka.co/blog/math-and-statistics-for-data-science/
  5. Lin, K. (2019, February 21). Role of Data Science in Artificial Intelligence. Retrieved January 08, 2021, from https://towardsdatascience.com/role-of-data-science-in-artificial-intelligence-950efedd2579
  6. What is AI? Artificial Intelligence Tutorial for Beginners. (n.d.). Retrieved January 08, 2021, from https://www.guru99.com/artificial-intelligence-tutorial.html