Using Insights, Anomaly Detection, and Forecasts

Insights, anomaly detection, and forecasts can help you make informed decisions with minimal effort by providing information about data trends and patterns, outliers, key drivers, and data metric changes over time. The information can be presented as natural-language summaries that you can easily customize.

Note: This function is only available with the Self-Service feature. If you want to purchase this feature, contact your Ricoh representative.

The built-in machine learning algorithm that is used for anomaly detection and forecasting continuously analyzes your data to identify outliers, to determine patterns and trends, and to reliably predict changes for key data metrics.

To begin using the machine learning capabilities, the dataset must meet these requirements:

  • The dataset must include at least one metric, such as printed jobs, printer throughput, or ink usage data, and at least one category dimension, such as printer names, locations, or operator names. Categories with NULL values are ignored.

    If you want to analyze anomalies or forecasts, you also need at least one date dimension.

  • Anomaly detection requires a minimum of 15 data points for training. For example, if the grain of your data is daily, you need at least 15 days of data. If the grain is monthly, you need at least 15 months of data.
  • Forecasting works best with more data. Make sure that your dataset has enough historical data for optimal results. For example, if the grain of your data is daily, you need at least 38 days of data. If the grain is monthly, you need at least 43 months of data. These are the requirements for each time grain:
    • Years: 32 data points
    • Quarters: 35 data points
    • Months: 43 data points
    • Weeks: 35 data points
    • Days: 38 data points
    • Hours: 39 data points
    • Minutes: 46 data points
    • Seconds: 46 data points

In this section:

  1. Using Insights
    Insights include useful summaries that can help your decision-making process by providing a better understanding of your data. They include descriptive text instead of charts and can display easy-to-follow key points and conclusions based on your data. For example, insights can provide information about growth rate, top and bottom movers, top and bottom ranked categories, metric changes over time, or minimum and maximum values. They can also include anomaly detection and forecasts.
  2. Configuring Anomaly Detection
    Anomaly detection uses the built-in machine learning algorithm to discover hidden trends and outliers in your data without the need for manual analysis or custom development. Anomalies are data points that fall outside an overall pattern of distribution. Identifying the anomaly causes and correlations can help you make data-driven decisions.
  3. Adding Forecasts in Visuals
    The built-in machine learning algorithm lets you easily forecast key data metrics. For example, you can forecast printer throughput to see if you can meet your goal by the end of the year. Anomalies in the data are excluded so they do not influence the forecast. You can also create interactive what-if analyses to determine the growth trajectory that you need to meet goals.
Parent topic: Creating and Managing Visuals