Activities

Dive into the Activities dashboard in Alviss AI, your essential tool for visualizing and analyzing raw data from active datasets. Track metrics like sales, investments, and trends through dynamic overviews, graphs, and visuals. Learn key features, best practices for data validation, filtering, exploratory analysis, and team collaboration to identify patterns and ensure reliable insights.

Overview

The Activities dashboard is your primary tool for visualizing and analyzing raw data within Alviss AI. It generates dynamic overviews for each file in your active Dataset, enabling you to track essential metrics like total sales, latest investments, or engagement trends. This dashboard supports exploratory data analysis (EDA), helping identify patterns, anomalies, and relationships before advancing to modeling or insights.

By default, Activities uses the current [Active Dataset](./Data/Datasets#Active Dataset). To ensure consistency between Activities and the Effect dashboard, adjust the Activities - Dashboard Data Source in your project settings.

Key Features

  • Overview Dashboards: Automatically created for every file in your Dataset, these summarize key data points, trends, and aggregates (e.g., total and latest values for sales or investments).
  • Graphs and Visuals: Includes a range of charts, such as line graphs for time series, bar charts for comparisons, and heatmaps for correlations, making EDA intuitive and efficient.
  • Tooltips: Hover over any graph, metric, or element for detailed explanations, contextual info, and raw values, enhancing understanding without cluttering the view.

Remember to use Filtering for more accurate data analysis, such as isolating specific regions, products, or time periods.

Best Practices

  • Data Validation: Before proceeding to model building, inspect for outliers, missing values, or inconsistencies here. Use visuals to spot trends and tooltips for deeper dives.
  • Consistency Settings: If your workflow involves both Activities and Effect dashboards, update project settings to use the same Dataset, ensuring aligned insights.
  • Exploratory Workflow: Start with broad overviews, then apply filters to drill down. Combine with Data Explorer for variable-specific analysis.
  • Team Collaboration: Share dashboard views or export visuals for reports, leveraging team access for collaborative reviews.

Before building models, it's important to thoroughly check your data for any outliers or missing observations. Ensuring data quality will improve the accuracy and reliability of your results.