Data Analytics TRAINING MODULE 1 PDF

# Data Analytics ## What is Data Analytics? ### Introduction - A diagram shows **Data Mining** in the center of a Venn Diagram, overlapping with **Machine Learning** and **Artificial Intelligence**. - **Statistics** is the third part of the Venn Diagram, overlapping only with **Data Mining**. -...

# Data Analytics ## What is Data Analytics? ### Introduction - A diagram shows **Data Mining** in the center of a Venn Diagram, overlapping with **Machine Learning** and **Artificial Intelligence**. - **Statistics** is the third part of the Venn Diagram, overlapping only with **Data Mining**. - An image of a human head with a blurred background of code and data is shown near the Venn Diagram. ### Why Data Analytics? - Several logos are presented in a row representing popular businesses that use data analytics to their advantage. - The logos represent **Facebook**, **Amazon**, **Uber**, **Bank of America**, **Airbnb**, and **Spotify**. - The logos are accompanied by a phrase describing how these platforms use data analytics. - **Facebook:** Social Analytics - **Amazon:** Improving E-commerce experience - **Uber:** Optimizing Rides - **Bank of America:** Increasing Customer experience - **Airbnb:** Improving Searches - **Spotify:** Music Recommendation ### Data - Data is defined as a set of values of qualitative or quantitative variables. - It is information in raw or unorganized form. - Data can be a fact, figure, characters, or symbols. - Information is defined as organized or meaningful data. - Information comes from analyzing data. ### What is Data Analytics? - Analytics is the discovery, interpretation, and communication of meaningful patterns in data. - Data Analytics (DA) is the process of examining data sets to draw conclusions about the information. - Analytics is not a tool or technology, rather it is the way of thinking and acting on data. - A caption below a diagram of hands holding a graph states: "Data on its own is useless unless you can make sense of it!" ### Primary Focus Areas for Analytics - A diagram shows a two by two quadrants of a grid. - The quadrants represent focus areas within data analytics. - The X-axis represents **Understand Customers Behavior** on the left and **Understand Products Usage** on the right. - The Y-axis represents **Increase Operational Efficiency** on the bottom and **Business Models Innovation** on the top. ### Data Analytics - Example - **Business Analytics:** - The business environment today is more complex than ever before. - Businesses are expected to be diligently responsive to the increasing demands of customers, various stakeholders, and even regulators. - The primary objective of an organization that seeks to turn to analytics is: - Revenue/Profit growth - Optimize expenditure - Organizations are turning to analytics to enhance their competitiveness. - More than 83% of Global CIOs surveyed by IBM in 2010 saw Business Intelligence and Analytics as a visionary plan to enhance their competitiveness. - **Risk** - **Fraud** - **Health** - **Web** ### The Process of Statistical Analysis - A diagram shows a process flow of three steps. - **Form Hypotheses** - **Identify Data Source** - **Prove/Disprove Hypothesis** - Statistical analysis enables us to make quantitative inferences based on an amount of information we can analyze (a sample). ### Data Analytics Life Cycle - A diagram shows a circular diagram of steps within the Data Analytics Life Cycle. - **Data Visualization** - **Data Analytics** - **Objective** - **Understanding the data** - **Data cleaning & data transformation** - **Data Enhancement** ### Types of Data Analytics - **Descriptive Analytics** - **Predictive Analytics** - **Prescriptive Analytics** ## Types of Data Analysis - A triangle diagram shows **Prescriptive Analytics** at the peak. **Predictive Analytics** and **Descriptive Analytics** are at the bottom corners. - **Prescriptive Analytics** - Enabling smart decisions based on data - What should we do? - **Predictive Analytics** - Predicting the future based in historical patterns - What could happen? - **Descriptive Analytics** - Mining data to provide business insights - What has happened? ## Types of Data Analysis - **Descriptive Analytics** - Aims to uncover valuable insights from data being analyzed. - Answers the question "What has happened?" - **Predictive Analytics** - Helps forecast behavior of people and markets. - Answers the question "What could happen?" - **Prescriptive Analytics** - Suggests conclusions or actions to be taken based on analysis. - Answers the question "What should be done?" - What should we do? - A set of images shows examples of each type of data analytics. - A picture shows Netflix recommending movies based on personal preferences. - A picture shows a grocery cashier handing out coupons. - A picture shows an airplane ticket with different prices. - Descriptive Analytics: Insight into the past. - Predictive Analytics: Understanding the future. - Prescriptive Analytics: Advice on possible outcomes. ## Descriptive Data Analytics - Although it is the simplest type of data analysis, it is the most commonly used. - **Types of Descriptive Analysis:** - **Measures of Central Tendency:** Tells us about the middle - Mean: The average - Median: The midpoint of the responses - Mode: The response with the highest frequency - **Measures of Dispersion:** - Range: The min, the max, and the distance between the two. - Variance: The average degree to which each of the points differ from the mean. - Standard Deviation: The most common way of expressing the spread of data - A histogram shows the **Mean**, **Median**, and **Mode** of the amount of items purchased. - A table shows data for **Customer ID**, **Items Purchased**, and **Amount Spent**. ## Variability - A diagram shows two groups of money bags. - The first group shows all money bags of equal size with a caption stating **No Variability in Cash Flow**. - The second group shows money bags with different sizes with a caption stating **Variability in Cash Flow**. - Both groups show a single money bag at the end representing the **Mean**. ## Standard Deviations - An image shows the **Standard Deviation** of a bell curve. - 68% within 1 Standard Deviation - 95% within 2 Standard Deviations - 99.7% within 3 Standard Deviations ## Predictive Data Analytics - Some mistake predictive analysis to only have relevance to predicting future events. - However, in cases such as sentiment analysis, existing data (e.g., the text of a tweet) is used to predict non-existent data (whether the tweet is positive or negative). - Several of the models that can be used for predictive analysis are: - Forecasting - Simulation - Regression - Classification - Clustering ## Prescriptive Data Analytics - Decisions can be formulated from descriptive and predictive analysis. - If I need to cut a product and I know that product C is least preferred and least profitable, I will cut product C. - Prescriptive Analytics explicitly tells you the decisions you should make. - This can be done using various techniques: - Linear Programming - Integer Programming - Mixed Integer Programming - Nonlinear Programming ## Comparing The Three Types of Data Analytics - Descriptive analysis is the most common. - A good practice is performing descriptive analyses prior to prescriptive/predictive analysis. - Understand that distribution, variance, skew, etc., may exclude certain models. - **How to know which type of analysis to pursue:** - How much time do you have? - What resources are available to you? - How accurate is your data? - How accurate do you need the model/analysis to be? - How popular/accepted is the model you are considering? - Don't subscribe to "that's how we've always done it," but remember to use a model that stakeholders will accept. - A diagram shows a graph with **Difficulty** on the X-axis and **Value** on the Y-axis. - **Hindsight** is displayed along the bottom (low value) of the graph, representing **Descriptive Data Analytics**. - **Insight** is displayed in the middle of the graph, representing **Predictive Data Analytics**. - **Foresight** is displayed at the top (high value) of the graph, representing **Prescriptive Data Analytics**. ## Why Big Data Analytics? - A diagram shows three overlapping circles. - **Big Data Analytics** is in the center. - **Cost reduction**, **Faster, better decision making**, and **New products & services** are displayed in the other circles, respectively. - Benefits of using big data analytics: - Helps organizations harness data and use it to identify new opportunities. - Leads to smarter business moves. - Results in more efficient operations. - Creates higher profits and happier customers. - A diagram shows a circle with **Data Analytics** in the center. - Arrows point from the circle with various phrases. - **Predict** - **Describe** - **Optimise** - **Trust** - **Discover** - **Empower** - **Embed** - **Delivering Change through Analytics** ## Data Analytics tools - Logos for the following data analytics tools are presented. - **kissmetrics** - **Tableau** - **Keen IO** - **HEAP** - **Google Analytics** - **crazyegg** ## Data Analyst/Journalist - A picture of magnifying glass on a bar graph is presented. - **Data Analyst/Journalist** role is described thus: - Research and extract information from various sources. - Describe the main features of a data set in support of business operations and decision making. ## Data Scientist - A beaker containing graphs and data is presented in an image. - **Data Scientist** role is described thus: - Cleanse data. - Apply various techniques to large data sets and create visualizations to understand abstract business questions. - Provide predictive insights into the future. - Anticipate future behaviors ## Data Scientist: The Sexiest Job in the 21st Century - A large text caption reads **Data Scientist: The Sexiest Job in the 21st Century**. - A business analyst isn't able to discover insights from huge sets of data. - Data scientists can work in coordination with different verticals within an organization. - Data Scientists find useful patterns and insights for a company to make tangible business decisions. - **Between 2011-2012, there was a 15,000% increase in job postings for data scientists in the US.** ## Use Cases - The logos from the first page are presented again. - Each logo is accompanied by a phrase describing how these platforms use data analytics. - **Facebook:** Social Analytics - **Amazon:** Improving E-commerce experience - **Uber:** Optimizing Rides - **Bank of America:** Increasing Customer experience - **Airbnb:** Improving Searches - **Spotify:** Music Recommendation ## Facebook - **2.5 Billion** monthly active users - **Deep learning** is utilized, including facial recognition and text analysis. - **Facebook uses powerful neural networks to classify faces in photographs.** - **Deep Text** engine is used to understand user sentences. - **Deep learning** is used for targeted advertising. - Facebook uses insights gained from data to cluster users based on their preferences and provide advertisements that appeal to them. - A picture of a social network graph is presented, with nodes marked as **User V**, **User W**, **User X**, **User Y**, and **User Z**. ## Amazon - Amazon heavily relies on **predictive analytics** to increase customer satisfaction. - Amazon uses a **personalized recommendation system**. - Amazon draws suggestions from other users who used similar products or provided similar ratings. - A diagram shows a process flow for **Amazon's Recommendation Engine**. - **Amazon S3:** Storing user data and inventory information. - **Amazon Personalize API:** Using personalize API to create a flow of user activity. - **Amazon Personalize:** Processing and examining the data, identifying the meaningful, selecting correct algorithms, and training and optimizing a personalization model for customizing the data. - **Loading Data:** - **Data Inspection:** - **Feature Identification:** - **Selecting the Right Algorithm:** - **Selecting the hyper parameters:** - **Trainingthe models:** - **Optimizingthe models:** - **Building an assembly of features:** - **Model Hosting:** - **Creation of the real-time caches:** -- **Customized Personalization API:** Using the user activity stream to produce recommendations. ## Amazon - **Fraud Detection** - Amazon has its own novel ways and algorithms to detect fraud sellers and fraudulent purchases. ## Uber - Uber is a popular smartphone application that allows you to book a cab. - Uber makes extensive use of **Big Data**. - Uber maintains a large database of drivers, customers, and several other records. - Whenever you hail for a cab, Uber matches your profile with the most suitable driver. - A list outlines the fees charged by Uber: - **Base (or initial) fare:** A flat fee charged at the beginning of a ride. - **Cost per minute:** How much is charged for each minute in the ride. - **Cost per mile:** How much is charged for each mile of the ride. - **Booking Fee (Formerly 'Safe Rides Fee'):** A flat fee to cover Uber's operating costs (Not included for Uber's more luxury services like UberBlack or UberSUV) - An equation is presented to calculate the Uber fare: $Base Fare + (Cost per minute * time in ride) + (Cost per mile * ride distance) + Booking Fee = Your Fare$. - The Uber logo is presented. ## Bank of America - Using Data to Leverage Customer Experience - Bank of America makes use of **Data Science** and **predictive analytics**. - Banking industries are able to detect frauds in payments and customer information. - Bank of America prevents frauds regarding insurances, credit cards, and accounting. - Banks employ **data scientists** to use their quantitative knowledge where they apply algorithms like association, clustering, forecasting, and classification. - **Risk modeling** using **Machine Learning**. - Bank of America seeks to minimize risk modeling. - The Bank of America logo is presented. ## Bank of America - Using Data to Leverage Customer Experience - Bank of America utilizes **intelligent customer segmentation** to understand their customers. - This segmentation categorizes customers into **high-value** and **low-value** segments. - **Clustering**, **logistic regression**, and **decision trees** are used by banks to understand the **Customer Lifetime Value (CLV)** and take group them in the appropriate segments. - The Bank of America logo is presented. ## Airbnb - Users can host accommodations and find accommodations through the Airbnb app and website. - Airbnb contains massive **big data** of customer and host information, stay records, and website traffic. - Users from certain countries would click the neighborhood link, browse the page and photos and not make any booking. - A different version of the app was released to mitigate this issue. - This version replaced neighborhood links with top travel destinations. - This resulted in a 10% improvement in the lift rate for users from those countries. - The Airbnb logo is presented. ## Spotify - Spotify is an **online music streaming giant**. - Spotify deals with a massive amount of **big data** from over 100 million users. - Spotify uses the 600 GBs of daily data generated by users to build algorithms to **boost user experience**. - In 2017, Spotify used **data science** to gain insights about which universities had the highest percentage of party playlists and which ones spent the most time on it. - The Spotify logo is presented.