IT in Business Data Science Applications PDF

Course: IT in Business Instructor: Hafsa Naeem Session 10 Data Science Applications Cloud Computing  DATA SCIENCE Data science is the application of data-centric computational and inferential thinking to understand the world and solve problems. Joseph Gonzalez, Professo...

Course: IT in Business Instructor: Hafsa Naeem Session 10 Data Science Applications Cloud Computing  DATA SCIENCE Data science is the application of data-centric computational and inferential thinking to understand the world and solve problems. Joseph Gonzalez, Professor at UC Berkeley.  Real-Life Example: E-Commerce Data Pipeline Role Focus Example Scenario Activities and Outcomes Designing and - Implementing ETL processes for data collection. Ensuring Smooth Data maintaining an e- Data Engineers - Setting up optimized data storage. Flow commerce data - Outcome: Reliable data pipeline. pipeline. Analyzing e- - Analyzing user behavior for patterns. Extracting Value from commerce data for - Building machine learning models. Data Scientists Data and insights and - Outcome: Actionable insights and automated Implementing Models implementing predictions. models. Focusing on refining - Iteratively improving a statistical model to best fit Improving a Single Statisticians a specific statistical the data. Model model. - Outcome: Enhanced model accuracy. Translating business - Identifying key business questions and goals. Bridging Data and requirements into - Collaborating with stakeholders. Business Analysts Business Needs data-driven - Outcome: Actionable insights aligned with business solutions. objectives.  ALGORITHMS & MODELS An algorithm is a set of A data model organizes commands that must be data elements and followed for a computer standardizes how the to perform calculations data elements relate to or other problem-solving one another. operations. Machine Learning Model == Model Data + Prediction Algorithm  ALGORITHMS & MODELS Algorithms are methods or procedures taken in other to get a task done or solve a problem, while Models are well-defined computations formed as a result of an algorithm that takes some value, or set of values, as input and produces some value, or set of values as output.  ALGORITHMS & MODELS Analogy Explanation Algorithm: The recipe for baking a cake. It's a set of step-by-step instructions. Cake Baking Model: The finished cake, the tangible result of following the recipe. Relation: The recipe guides the process (algorithm), and the cake is the outcome (model). Algorithm: The instruction manual for a LEGO set, providing step-by-step guidance. Building a LEGO Set Model: The completed LEGO structure, the tangible result of following the instructions. Relation: The instructions (algorithm) lead to the creation of the LEGO model. Algorithm: The route calculation and guidance provided by the GPS system. Model: Your journey on the map, representing the guided path. GPS Navigation Relation: The GPS algorithm determines the route, and your journey is the model of that route. Algorithm: A recipe for cooking a specific dish, specifying ingredients and steps. Cooking a Dish Model: The prepared dish, the outcome of following the recipe. Relation: The recipe (algorithm) results in the dish (model) when followed correctly.  Data Science Workflow 1. Data Collection 2. Experimentation and Prediction 3. Exploration and Visualization  Data Science Workflow 1. Data Collection a. Customer surveys b. Web traffic results c. Emails between a sales team and potential clients d. Financial transactions. 2. Experimentation and Prediction 3. Exploration and Visualization  Applications of Data Science 1. Machine Learning (ML) 2. Internet of Things (IoT) 3. Deep Learning 4. Generative AI  MACHINE LEARNING 1. A well-defined question 2. A set of example data 3. A new set of data to use our algorithm on  FRAUD DETECTION CASE Case Study: Enhancing Fraud Detection at XYZ Bank Overview: As the leader of Fraud Detection at XYZ Bank, our objective is to employ advanced data analysis to assess the probability of a transaction being fraudulent. We initiated this process by meticulously gathering transaction details, including amount, date, transaction type, and cardholder address. Question: Our focus is on determining the probability of a transaction being fraudulent. Specifically, we inquire, "What is the probability that this transaction is fraudulent?" Data: We've compiled a robust dataset, including labeled examples of past credit card transactions marked as "fraudulent" or "valid." This dataset forms the foundation for training our fraud detection algorithm. Algorithm Application: Our well-trained algorithm is seamlessly applied to new credit card transactions, predicting the probability of fraud based on patterns learned from historical data. Adaptive Learning: With each new transaction, the algorithm refines its predictive capabilities, adapting to evolving patterns and ensuring continued effectiveness in identifying potential fraudulent activities. Insights: By posing a clear question, leveraging a curated dataset, and applying our algorithm to new transactions, XYZ Bank gains actionable insights. The algorithm provides probabilities of fraud, enabling proactive measures to safeguard our customers.  ML SOLUTION 1. "What is the probability that this transaction is fraudulent? 2. Old transactions labeled as"fraudulent" or "valid" 3. New credit card transactions  Internet of Things Gadgets which are not standard computers but still have the ability to transmit data  Internet of Things 1. Smart watches 2. Internet-connected home security systems 3. Electronic toll collection systems 4. Building energy management systems 5. Much, much more!  Smart Watches 1. Detect different physical activities like running or walking 2. An Accelerometer monitors motion in three dimensions 3. The data generated by the sensor is the basis of machine learning problem 4. You can develop the algorithm that can recognize and accelerometer data as representing one of two states, i.e., running or walking  Deep Learning Many neurons work together Requires much more training data Used in complex problems Image classication Language learning/understanding Eg., Self driving cars  GENERATIVE AI “Any sufficiently advanced technology is simply undistinguishable from Magic.” Arthur C. Clark  IMPORTANCE OF GEN AI We can access concise information in just a manner of seconds. We can also automatically generate text such as news articles or product Access descriptions. Automate We can even design custom products like Design shoes or furniture. Generate We can produce music, speech, visual effects, 3D assets and sound effects using algorithms trained on already existing data. IMPORTANCE OF GEN AI  WHAT IS GEN AI? The type of AI that generates new content. Definition Generative AI is specifically designed to generate new content as its primary output.  APPLICATIONS OF GEN AI Generative AI is often used in applications such as: Applications Image generation Video synthesis Language generation Music composition  Anomaly Detection Generative models can be trained on a specific dataset to learn the patterns and distribution of normal data. Data Augmentation Missing Data Imputation Generative models can be Generative models can be used to generate synthetic trained on the available data that closely resembles data to learn the underlying the real data distribution. patterns and relationships. GEN AI IN DATA SCIENCE  Recommendation Systems Generative models can be employed in recommendation systems to generate personalized recommendations for users. Synthetic Data Generation Exploratory Data Analysis Generative AI models, such Generative models can be as GANs, can generate new used to explore and synthetic data that closely understand the underlying resembles the original structure of a dataset. dataset. GEN AI IN DATA SCIENCE  CLOUD COMPUTING Cloud Computing provides us a means by which we can access the applications as utilities, over the Internet. It allows us to create, configure, and customize applications online. Concept With Cloud Computing users can access database resources via the internet from anywhere for as long as they need without worrying about any maintenance or management of actual resources.  What is Cloud? The term Cloud refers to a Network or Internet. In other words, we can say that Cloud is something, which is present at remote location. Cloud can provide services over public networks or on private networks, i.e., WAN, LAN Definition or VPN. Applications such as e-mail, web conferencing., Customer relationship management (CRM), all run in cloud.  What is Cloud Computing? Cloud Computing refers to manipulating, configuring, and accessing the applications online. It offers online data storage, infrastructure and application Definition Cloud Computing is both a combination of software and hardware based computing resources delivered as a network service.  Cloud Computing Models There are certain services and models working behind the scene making the cloud computing feasible and accessible to end users. Working Following are the working models for cloud Models computing: 1. Deployment Models 2. Service Models  Deployment Models Deployment models define the type of access to the cloud, i.e., how the cloud is located? Cloud can have any of the four types of access: Deployment 1. Public Models 2. Private 3. Hybrid 4. Community  Deployment Models PUBLIC CLOUD : The Public Cloud allows systems and services to be easily accessible to the general public. Public cloud may be less secure because of its openness, e.g., Email. PRIVATE CLOUD: The Private Cloud allows systems and services to be accessible within an organization. It offers increased security because of its private nature. Deployment COMMUNITY CLOUD : The Community Cloud allows systems Models and services to be accessible by group of organizations. HYBRID CLOUD : The Hybrid Cloud is mixture of public and private cloud. However, the critical activities are performed using private cloud while the non-critical activities are performed using public cloud.  Service Models Service Models are the reference models on which the Cloud Computing is based. These can be categorized into three basic service Service models as listed below: Models 1. Infrastructure as a Service (IaaS) 2. Platform as a Service (PaaS) 3. Software as a Service (SaaS)  Infrastructure as a Service (IaaS) IaaS is the delivery of technology infrastructure as an on demand scalable service. IaaS provides access to fundamental resources such as physical machines, virtual machines, IaaS virtual storage, etc Usually billed based on usage Usually multi tenant virtualized environment Can be coupled with Managed Services for OS and application support  Infrastructure as a Service (IaaS) Amazon Web Services (AWS) Microsoft Azure Google Cloud Platform (GCP) IBM Cloud Infrastructure DigitalOcean OpenStack  Platform as a Service (PaaS) PaaS provides the runtime environment for applications, development & deployment tools, etc. PaaS provides all of the facilities required to support the complete life cycle of building and delivering web applications and services entirely PaaS from the Internet. Typically applications must be developed with with a a particular platform in mind Multi tenant environments Highly scalable multi tier architecture  Platform as a Service (PaaS) Amazon Web Services (AWS) Google App Engine Microsoft Azure App Service Red Hat OpenShift Salesforce Lightning Platform IBM Cloud Foundry  Software as a Service (SaaS) SaaS model allows to use software applications as a service to end users. SaaS is a software delivery methodology that provides licensed multi-tenant access to software SaaS and its functions remotely as a Web-based service. Usually billed based on usage Usually multi tenant environment Highly scalable architecture  Software as a Service (SaaS) Salesforce Google Workspace (formerly G Suite) Microsoft 365 Dropbox Slack Zoom Shopify Netflix Adobe Creative Cloud Salesforce Marketing Cloud  Cloud Storage Create an account on Cloud Upload your files, content Log on to your account to access your content over WiFi Content stays on cloud  ADVANTAGES & DISADVANTAGES Lower computer costs Requires a constant Improved performance Internet connection Reduced software costs Does not work well with Instant software updates low-speed connections Improved document format Features might be compatibility limited Unlimited storage capacity Can be slow Increased data reliability Stored data can be lost Universal document access Stored data might not be Latest version availability secure Easier group collaboration Cloud Storage Device independence THANKS!