Data Science project 1. Customer Segmentation: Objective: Use clustering algorithms to segment customers based on their purchasing behavior. Data: Customer purchase history, demographic information, etc. Tools: Python, pandas, scikit-learn, matplotlib, seaborn. 2. Time Series Forecasting for Stock Prices: Objective: Predict future stock prices using time series analysis. Data: Historical stock price data. Tools: Python, pandas, NumPy, scikit-learn, matplotlib, seaborn. Machine Learning project1. Sentiment Analysis for Product Reviews: Objective: Analyze customer reviews to determine the sentiment towards a product. Data: Customer reviews from websites like Amazon, Yelp, etc. Tools: Python, pandas, scikit-learn, NLTK, matplotlib, seaborn. 2. Image Recognition with Neural Networks: Objective: Build a neural network to recognize images. Data: Image datasets like CIFAR-10, MNIST, etc. Tools: Python, TensorFlow, Keras, OpenCV. Artificial Intelligence project1. Chatbot Development: Objective: Develop a chatbot that can handle specific tasks or answer questions. Data: Conversational data, domain-specific data. Tools: Python, Rasa, TensorFlow, Dialogflow. 2. Game AI Development: Objective: Develop an AI that can play and possibly master a specific game. Data: Game data, player data. Tools: Python, Unity, TensorFlow, PyTorch. Cloud project1. Serverless Image Processing Pipeline: Objective: Create a serverless pipeline to process images uploaded to a cloud storage bucket. Data: Images. Tools: AWS Lambda, Google Cloud Functions, Azure Functions, Python, OpenCV. 2. Cloud-based Machine Learning Model Deployment: Objective: Deploy a machine learning model in the cloud and expose it as a REST API. Data: Any data relevant to the machine learning model. Tools: AWS SageMaker, Google AI Platform, Azure Machine Learning, Docker, Python, Flask.

Paginated Report Visual in Power BI Visual Pane * Ease of Use: Easy to integrate within a Power BI report, providing a seamless experience. * Purpose: Primarily used to add paginated report-like visuals to your interactive Power BI reports. * Interactivity: Works within the Power BI ecosystem, allowing you to have interactive dashboards along with paginated visuals. * Data Source: Relies on the same dataset as the Power BI report. * Design Flexibility: Limited compared to Report Builder; you're restricted to the visual's capabilities within the Power BI interface. * Sharing and Deployment: Shares the same sharing and deployment options as Power BI reports. Paginated Reports Built through Report Builder* Ease of Use: Requires a separate tool (Report Builder) for creating reports, and then uploading to the Power BI service. * Purpose: Designed for scenarios where you need traditional, pixel-perfect, paginated reports that can be printed or shared as PDFs. * Interactivity: Limited interactivity compared to Power BI reports; mainly used for static, printable reports. * Data Source: Can connect to a variety of data sources, including those not available in Power BI. * Design Flexibility: Offers more design flexibility, with detailed control over the layout, formatting, and grouping of data. * Sharing and Deployment: Can be shared and deployed through the Power BI service, but also through other means like email subscriptions.

Sales Header = IF ( HASONEVALUE ( Subsidiary[Name] ), "Sales for " & VALUES ( Subsidiary[Name] ), "Total Sales" ) Cost Header = IF ( HASONEVALUE ( Subsidiary[Name] ), "Cost for " & VALUES ( Subsidiary[Name] ), "Total Cost" ) GP Header = IF ( HASONEVALUE ( Subsidiary[Name] ), "GP for " & VALUES ( Subsidiary[Name] ), "Total GP" )

To set up incremental refresh for monthly snapshot data: 1. Parameters: * In Power Query Editor, set up two date parameters: RangeStart and RangeEnd. * These parameters will be used to filter the data that gets loaded into the model. 1. Filtering: * Still in Power Query Editor, filter your history table using these parameters. For example, you might use a filter like: Date >= RangeStart and Date < RangeEnd. 1. Incremental Refresh Configuration: * Close the Power Query Editor and go to the table's properties in the Fields pane. * Configure the incremental refresh: * Set the "Store rows in the last" option to the number of months of data you want to keep in the model. For example, if you want to keep 12 months of data, set it to 12 months. * Set the "Refresh rows in the last" option to 1 month, since you're adding data monthly. * Use the RangeStart and RangeEnd parameters appropriately. 1. Publish to Power BI Service: * Incremental refresh is a feature of the Power BI Service, so you'll need to publish your .pbix file to the service. * Once published, configure the dataset's scheduled refresh. The Power BI Service will use the incremental refresh settings to only load the new month's data on each refresh. 1. Handling the Base Table: * For your requirement of making the earliest month a base table and the remaining months as incremental, you can initially set the RangeStart and RangeEnd to cover just the earliest month when you first publish the dataset. This will load only that month's data. * Subsequently, when you adjust the parameters for scheduled refresh in the Power BI Service, set them to cover all months from the earliest month to the current month. This way, the earliest month's data will remain as the base, and only new months' data will be added incrementally.

Creating a Purchase Requisition (PR) in SAP 1. On the SAP Easy Access screen, enter transaction code ME51N and press Enter. 2. In the Purchase Requisition screen, enter the necessary details such as Document Type, Material, Quantity, Plant, etc. 3. Click on the check mark button to check for any errors. 4. If there are no errors, click the Save button to save the purchase requisition. A message will appear at the bottom of the screen with the PR number. Creating a Purchase Order (PO) in SAP1. On the SAP Easy Access screen, enter transaction code ME21N and press Enter. 2. In the Purchase Order screen, enter the necessary details such as Document Type, Vendor, Purchasing Organization, Purchasing Group, etc. 3. In the item overview section, enter the Material, Quantity, Plant, etc. 4. Click on the check mark button to check for any errors. 5. If there are no errors, click the Save button to save the purchase order. A message will appear at the bottom of the screen with the PO number.

Here's a step-by-step guide to create a "% of Parent Row Total" measure for a multi-level hierarchy: 1. Sample Data:Assuming you have a table named Sales with the following columns: * Category * SubCategory * SalesAmount 2. Create a Hierarchy:In the Fields pane, drag the Category field above the SubCategory field to create a hierarchy. Name this hierarchy as CategoryHierarchy. 3. Create the Measure:To calculate the "% of Parent Row Total", you'll need to determine the context of the current row and then calculate the percentage based on the parent's total. Here's a DAX formula to achieve this: % of Parent Row Total = VAR CurrentLevel = SELECTEDVALUE('Sales'[CategoryHierarchy], BLANK()) VAR ParentTotal = SWITCH(TRUE(), ISBLANK(CurrentLevel), SUM('Sales'[SalesAmount]), // If top level, take total sales VALUES('Sales'[Category]), // If subcategory level, take category total BLANK() // Default ) RETURN DIVIDE(SUM('Sales'[SalesAmount]), ParentTotal) 4. Use the Measure in a Matrix Visualization: * Drag the CategoryHierarchy to the Rows shelf of a Matrix visualization. * Drag the SalesAmount and % of Parent Row Total measures to the Values shelf.

1. Customer Segmentation Objective: Use clustering algorithms to segment customers based on their purchasing behavior. Steps: * Collect or find a dataset that includes customer purchase history and possibly demographic information. * Perform exploratory data analysis (EDA) to understand the data. * Preprocess the data, handling any missing values and outliers. * Use clustering algorithms like K-means, hierarchical clustering, or DBSCAN to segment the customers. * Analyze the results and create visualizations to show the different customer segments. * Interpret the results and propose strategies to target each customer segment. 2. Time Series Forecasting for Stock PricesObjective: Predict future stock prices using time series analysis. Steps: * Collect historical stock price data. * Perform EDA to understand the trends and patterns in the data. * Preprocess the data, handling any missing values and outliers. * Use time series forecasting models like ARIMA, SARIMA, or LSTM to predict future stock prices. * Evaluate the model's performance using appropriate metrics. * Visualize the predictions and compare them to the actual stock prices. 3. Sentiment Analysis for Product ReviewsObjective: Analyze customer reviews to determine the sentiment towards a product. Steps: * Collect customer reviews from websites like Amazon, Yelp, etc. * Perform EDA to understand the data. * Preprocess the text data, handling any missing values and outliers. * Use natural language processing (NLP) techniques to extract features from the text. * Use machine learning models like Naive Bayes, SVM, or deep learning models like LSTM to predict the sentiment. * Evaluate the model's performance using appropriate metrics. * Visualize the results and create a dashboard to show the sentiment analysis results. 4. Serverless Image Processing PipelineObjective: Create a serverless pipeline to process images uploaded to a cloud storage bucket. Steps: * Set up a cloud storage bucket to store the images. * Create a serverless function that gets triggered when an image is uploaded to the bucket. * Write the code for the serverless function to process the image (e.g., resize, filter, etc.). * Test the pipeline by uploading images to the bucket and checking the processed images. * Monitor the performance and costs of the serverless pipeline.