From Problem to Approach

From Problem to Approach Computing Grade 11 What is methodology and why does data science need it Driving Question 3 Lesson objectives ▰ Why we are interested in data science. ▰ What a methodology is, and why data scientists need a methodology. ▰ The data science methodology and its flowchart. ▰ How to apply business understanding and the analytic approach to any data science problem. 4 “ PDN In pairs, discuss what is your understanding of the word methodology. Which methodologies do you know? ” 5 Let us get started. Business Approach What is the problem that you are trying to solve? 7 “ Methodologies in a data science project ensure a systematic and effective approach to problem-solving, aligning the analysis with the goals and objectives of the business. ” 8 Why should we use methodologies in data science The Importance of Methodologies in Data Science ▰ Clarification and understanding: It's crucial to clarify the problem and understand the business goals before diving into data analysis. ▰ Defined questions: Clearly defining the questions helps focus the analysis on solving the actual problem, rather than addressing a perceived one. ▰ Goals and objectives: Establishing goals and breaking them down into objectives helps prioritize and organize the problem-solving approach. 9 The Importance of Methodologies in Data Science 10 ▰ Stakeholder involvement: Engaging stakeholders helps gather requirements and gain a comprehensive understanding of the problem. ▰ Prioritization and organization: Structured discussions and identifying key factors contribute to efficient planning and organization. ▰ Involvement of business sponsors: Key business sponsors provide guidance and support throughout the project, ensuring its success. ▰ Identifying requirements: Clearly identifying the specific business requirements guides the development of the appropriate model. Analytic Approach How can you use data to answer the question 12 “ This is the second stage of the data science methodology. Once a strong understanding of the question is established, the analytic approach can be selected. ” 13 Analytical Approaches - DDPP Descriptive ● Diagnostic Current status ● ● ● What if these trends continue? What will happen next? Predictive ● What happened? Why is it happening? How do we solve it Prescriptive DDPP Explained - Descriptive Analytics ▰ Descriptive analytics focuses on understanding historical data and providing insights into what has happened in the past. ▰ It involves summarizing and visualizing data to identify patterns, trends, and relationships. ▰ Descriptive analytics helps answer questions like "What happened?" and provides a foundation for further analysis. 14 DDPP Explained - Diagnostics Analytics ▰ Diagnostic Analytics: Diagnostic analytics goes beyond describing what happened and aims to understand why something happened. ▰ It involves digging deeper into the data to uncover the underlying causes or factors contributing to a particular outcome or event. Diagnostic analytics helps answer questions like "Why did it happen?" and helps in root cause analysis. 15 DDPP Explained - Predictive Analytics ▰ Predictive analytics uses historical data and statistical modeling techniques to make predictions or forecasts about future events or outcomes. ▰ It involves analyzing patterns and relationships in the data to develop predictive models. Predictive analytics helps answer questions like "What is likely to happen?" and enables proactive decision-making based on anticipated outcomes. 16 DDPP Explained - Prescriptive Analytics ▰ Prescriptive analytics takes the insights from descriptive, diagnostic, and predictive analytics a step further by providing recommendations or actionable insights. ▰ It leverages optimization algorithms, simulations, and other techniques to suggest the best course of action to achieve a desired outcome. ▰ Prescriptive analytics helps answer questions like "What should we do?" and assists in making informed decisions to optimize results. 17 18 Activity - Individual Work 1. State whether following scenarios in the slides to follow are Descriptive, Diagnostic, Predictive or Prescriptive analysis. 2. Write all your answers in your workbook and give a reason for your answers. 19 Scenario 1 A company's website experienced a sudden drop in user engagement. To diagnose the issue, the marketing team analyzes website data, user behavior patterns, and performs A/B testing. They discover that a recent change in the website's layout led to confusion among users, resulting in decreased engagement. By identifying the root cause of the problem, the team can take corrective actions to improve the website's performance. 20 Scenario 2 A logistics company wants to optimize its delivery routes to minimize costs and improve efficiency. Using prescriptive analytics, they integrate real-time data on traffic conditions, weather forecasts, and package volumes. Advanced algorithms suggest the most optimal routes, considering various factors, such as time, fuel consumption, and delivery deadlines. This enables the company to make data-driven decisions on the best course of action for route planning and resource allocation. 21 Scenario 3 An insurance company wants to predict the likelihood of a customer making a claim in the next year. They analyze historical customer data, including demographic information, previous claim history, and policy details. By applying predictive modeling techniques, they develop a model that predicts the probability of a customer filing a claim. This helps the company proactively manage risk and allocate resources accordingly. 22 Scenario 4 A retail store wants to understand the sales performance of its products over the past year. They analyze historical sales data, create visualizations, and generate reports to identify which products were the top sellers, the sales trends over different time periods, and the peak sales months. This helps the store gain insights into what happened in the past. 23 Exit Ticket 321 Activity 1. Think of three examples of how data analytics can be applied in different industries or domains. 2. Identify two benefits of using data analytics in decision-making. 3. Share one example of how data analytics can make a positive impact in your daily life. 24 References www.cognitiveclassroom.ai