Elective C1 Reviewer PDF

INTRODUCTION TO BUSINESS ANALYTICS performance and is typically designed using a relational BUSINESS ANALYTICS database management system (RDBMS). - involves using data, statistical methods, and p...

INTRODUCTION TO BUSINESS ANALYTICS performance and is typically designed using a relational BUSINESS ANALYTICS database management system (RDBMS). - involves using data, statistical methods, and predictive 4. METADATA modeling to inform business decisions Metadata provides information about the data, such as - It leverages various software tools to analyze data and generate definitions, source systems, data types, and transformation insights that can drive strategy, optimize operations, and predict rules. It helps users understand and navigate the data future trends warehouse. 5. DATA MARTS 9 Business Analytics Importance In Modern Business These are subsets of the data warehouse that are tailored to the 1. Data-Driven Decision Making specific needs of different business units or departments, such 2. Competitive Advantage as marketing, finance, or sales 3. Customer Understanding and Personalization 6. OLAP (ONLINE ANALYTICAL PROCESSING) TOOLS 4. Operational Efficiency These tools allow users to perform complex queries, analysis, 5. Financial Performance and reporting on the data stored in the data warehouse. OLAP 6. Innovation and Product Development tools support multidimensional analysis, enabling users to slice 7. Regulatory Compliance and Risk Management and dice data across various dimensions. 8. Strategic Planning 9. Employee Performance and Management DATA SOURCES o POS Systems: Capture daily sales transactions, customer OBJECTIVE OF BUSINESS ANALYTICS purchases, and payment methods. Focus on improving decision-making, enhancing efficiency, driving o CRM Systems: Store customer profiles, loyalty program data, revenue growth, and supporting strategic and operational goals and interaction history. through data-driven insights o Inventory Management Systems: Track stock levels, orders, and supplier information. DATA WAREHOUSING o Online Sales Platforms: Capture online transactions, website - refers to the process of collecting, storing, and managing large traffic, and customer behavior volumes of data from different sources in a central repository, known as a data warehouse ETL PROCESS - The primary purpose of a data warehouse is to consolidate data The company uses an ETL tool to extract data from these from multiple sources into a single, unified system that can sources. The data is then transformed into a consistent format, with support reporting, analysis, and decision-making standardized units of measurement, consistent naming conventions, and removal of duplicates. Finally, the cleaned data is loaded into 4 KEY CHARACTERISTICS OF A DATA WAREHOUSE the data warehouse. 1. SUBJECT-ORIENTED Data warehouses are organized around key subjects, such as DATA WAREHOUSE DATABASE customers, sales, products, or financials, rather than around the The integrated data is stored in a central data warehouse organization's ongoing operations. This organization helps database, which is designed to support complex queries and analysis. simplify data access and analysis The database includes tables related to sales, customers, products, inventory, and suppliers 2. INTEGRATED Data from various sources (e.g., databases, spreadsheets, and OLAP TOOLS AND ANALYSIS external sources) are integrated into the data warehouse, Business analysts and managers use OLAP tools to perform various ensuring consistency in data formats, naming conventions, and analyses, such as: measurements o Sales Analysis: Analyzing sales trends by region, store, and product category to identify best-selling products and 3. NON-VOLATILE underperforming stores. Once data is entered into the data warehouse, it does not o Customer Segmentation: Segmenting customers based on change. This characteristic ensures that historical data is purchase history and loyalty program data to target marketing preserved for analysis over time campaigns more effectively. o Inventory Optimization: Analyzing inventory levels and 4. TIME-VARIANT turnover rates to optimize stock levels and reduce carrying Data in a data warehouse is typically stored with timestamps to costs. allow for historical analysis. This enables businesses to analyze o Historical Trends: Comparing current sales data with historical trends over time. data to identify seasonal trends and forecast future demand 6 COMPONENTS OF DATA WAREHOUSE OUTCOME 1. DATA SOURCES By implementing a data warehouse, the retail company These are the different systems, databases, and files from which gains a comprehensive and unified view of its business operations. data is extracted. Sources can include operational databases, This centralized data repository enables the company to make data- CRM systems, ERP systems, and external data providers. driven decisions, optimize inventory management, improve 2. ETL PROCESS (Extract, Transform, Load) customer targeting, and ultimately enhance profitability This process involves extracting data from various sources, transforming it into a consistent format, and loading it into the data warehouse. ETL tools are critical for data integration and quality control. 3. DATA WAREHOUSE DATABASE The central repository where the integrated, cleaned, and organized data is stored. This database is optimized for query TYPES OF BUSINESS ANALYTICS 1. DESCRIPTIVE ANALYTICS PRESCRIPTIVE ANALYTICS - the process of analyzing historical data to understand and - the branch of advanced analytics that goes beyond predicting summarize what has happened in the past future outcomes to recommending specific actions or strategies - It focuses on using data aggregation, reporting, and to achieve optimal results visualization techniques to identify patterns, trends, and - It uses a combination of optimization, simulation, decision relationships within the data analysis, and advanced machine learning techniques to suggest the best course of action among various alternatives 4 KEY POINTS OF DESCRIPTIVE ANALYTICS - The main objective of prescriptive analytics is to guide 1. Historical Focus - Descriptive analytics is concerned with decision-makers in choosing the most effective strategies to analyzing past data to gain insights into past events. achieve desired business outcomes while considering 2. Data Summarization - It involves summarizing and organizing constraints and potential risks data to make it understandable and actionable. 3. Reporting and Visualization - Descriptive analytics often presents KEY COMPONENTS OF PRESCRIPTIVE ANALYTICS data in a visually accessible way through charts, graphs, dashboards, a. Optimization and reports. Optimization techniques are used to find the best possible 4. Decision Support - While descriptive analytics does not predict solution from a set of alternatives, given certain constraints future outcomes or prescribe actions, it provides a foundation for (such as budget, time, or resources). They help make the most informed decision-making by providing a clear picture of past efficient use of available resources to achieve the desired goals. performance. Example: A retailer may use optimization to determine the best inventory levels that minimize costs while meeting customer demand. PREDICTIVE ANALYTICS b. Linear Programming - is a branch of advanced analytics that uses historical data, - is a mathematical technique used in optimization that focuses statistical algorithms, and machine learning techniques to on maximizing or minimizing a linear objective function, predict future outcomes subject to linear equality and inequality constraints. - The primary goal of predictive analytics is to provide - It is widely used in prescriptive analytics to solve complex actionable insights that can inform decision-making by decision-making problems where the goal is to optimize a anticipating trends, behaviors, and events. certain objective, such as cost, profit, or resource utilization. Example: A company might use linear programming to KEY CONCEPTS IN PREDICTIVE ANALYTICS optimize its production schedule, determining how much of a. Regression Analysis - is a statistical method for examining the each product to produce in order to maximize profit while relationship between one dependent variable and one or more staying within budget and resource constraints. independent variables. It helps predict the value of the c. Simulation dependent variable based on the values of the independent - Simulation models allow decision-makers to assess the impact variables. of different decisions or scenarios by creating virtual models of Example: In retail, regression analysis can predict future sales real-world processes. based on factors like advertising spend, seasonality, and - These models help in understanding the potential outcomes and economic indicators. risks associated with various strategies. Example: A logistics b. Time Series Analysis - involves analyzing data points collected company might use simulation to test different routing or recorded at specific time intervals to identify trends, seasonal strategies and their impact on delivery times and costs. patterns, and cyclical behavior over time. Example: A company might use time series analysis to forecast SOFTWARE’S USED (EXCEL, PYTHON, R) IN BUSINESS monthly sales, taking into account past sales data, seasonal ANALYTICS trends, and external factors like holidays or economic cycles. A. EXCEL - One of the most widely used tools in business analytics, c. Modern Tools in Predictive Analytics: Predictive analytics is particularly for small to medium-sized data sets. It's favored for supported by various modern tools and platforms that facilitate its simplicity, ease of use, and wide accessibility. data analysis and model building. Common tools include: Key Features: o Python - With libraries like sci-kit-learn, TensorFlow, and o Data Analysis - Excel provides built-in functions and stats models, Python is widely used for building predictive formulas for basic statistical analysis, data cleaning, and models. manipulation. o R - A statistical programming language that offers powerful o Visualization - Excel offers various charting and graphing tools for data analysis, including regression and time series options, such as bar charts, line charts, and pivot charts, to analysis. help visualize data. o SAS and IBM SPSS - Commercial software solutions that o Pivot Tables - Pivot tables are a powerful feature in Excel provide a range of predictive analytics capabilities. that allows users to summarize, analyze, explore, and o Machine Learning Platforms - Cloud-based platforms like present data. Google AI, Microsoft Azure ML, and AWS SageMaker offer o Add-Ins - Excel supports add-ins like Power Query and robust environments for building and deploying predictive Power Pivot, which enhance its capabilities for data models analysis and modeling. o Scenario Analysis - Excel’s built-in tools like Goal Seek, d. Text Mining - Text mining involves extracting useful Solver, and Scenario Manager allow for optimization and information from unstructured text data using techniques such decision-making analysis. as natural language processing (NLP), sentiment analysis, and Common Use Cases: topic modeling. o Budgeting and financial forecasting Example: Companies might use text mining to analyze o Sales analysis and reporting customer reviews or social media posts to predict product o Basic statistical analysis and visualization satisfaction and potential issues. o Data cleaning and transformation PYTHON - A versatile, open-source programming language that is increasingly popular in business analytics due to its extensive libraries and ability to handle large data sets. Key Features: o Data Manipulation - Libraries like pandas and NumPy are widely used for data manipulation, cleaning, and transformation. o Data Visualization - Python has powerful libraries like Matplotlib, Seaborn, and Plotly for creating detailed and interactive visualizations. o Machine Learning and AI: Python's sci-kit, TensorFlow, and Keras libraries provide tools for building and deploying machine learning models. o Automation - Python can automate repetitive tasks and processes, such as data extraction, transformation, and loading (ETL). o Integration - Python can be integrated with databases, web services, and other software tools, making it highly flexible for complex workflows. Common Use Cases: o Advanced data analytics and modeling o Predictive analytics and machine learning o Data visualization and dashboard creation o Automating data processing pipelines R - A programming language and environment specifically designed for statistical computing and graphics. It's widely used in academia and industries that require advanced statistical analysis. Key Features: o Statistical Analysis - R has a vast collection of packages (e.g., ggplot2, dplyr, tidyverse) for statistical modeling, hypothesis testing, and data analysis. o Data Visualization - R’s ggplot2 package is renowned for creating complex and high-quality visualizations with ease. o Machine Learning - R offers several packages for machine learning, such as caret, randomForest, and xgboost, which are used for building predictive models. o Customizable Reports - R integrates with RMarkdown to create dynamic and reproducible reports that combine analysis, code, and narrative. o Extensibility - R can be extended through a wide variety of packages available from CRAN, the Comprehensive R Archive Network Common Use Cases: o Statistical analysis and hypothesis testing o Data mining and exploration o Advanced visualization and reporting o Predictive modeling and machine learning Each of these tools—Excel, Python, and R —plays a crucial role in business analytics. Excel is ideal for quick, straightforward analysis and reporting, while Python and R offer more advanced capabilities for handling large data sets, performing complex statistical analysis, and building predictive models. Depending on the specific needs of a business analytics project, these tools can be used individually or in combination to derive actionable insights from data