DataboxDatabox Databox

CQL (Concept Drift Quantification) is a metric that measures the amount of change in the underlying concept of data over time, enabling drift detection in machine learning models.

With Databox you can track all your metrics from various data sources in one place.

  • About
  • Technical Details
What is "CQL"?
CQL (Concept Drift Quality) is a metric used to measure the effectiveness of a concept drift detection algorithm. It quantifies how accurately the algorithm is able to detect drifts in the data. A high CQL score indicates that the algorithm is correctly identifying when the underlying concept in the data has shifted, making it a valuable tool for assessing the performance of a drift detection system.
Example: Detecting performance degradation in a key database cluster by monitoring the CQL metric.

Visualizations

  • Databox visualization

    Number

    Used to show a simple Metric or to draw attention to one key number.

How to track CQL in Databox?

Databox is a business analytics software that allows you to track and visualize your most important metrics from any data source in one centralized platform.

To track CQL using Databox, follow these steps:

  1. 1
    Connect Drift that contains the metric you want to track
  2. 2
    Select the metric you want to track from the list of available metrics
  3. 3
    Drag and drop the selected metric onto your dashboard
  4. 4
    Watch your dashboard populate in seconds
  5. 5
    Put CQL on the Performance screen
  6. 6
    Get CQL performance daily with Scorecards or as a weekly digest
  7. 7
    Set Goals to track and improve performance of CQL
Drift integration with Databox Track CQL from Drift in Databox GET STARTED

Drift CQL included in Dashboard Templates 1

  • Live view

    Drift Contacts & Conversations

    Drift dashboard template which gives you insights about new and closed conversations, reply times by team members and more.

    Drift

General

  • Description
    CQL (Concept Drift Quantification) is a metric that measures the amount of change in the underlying concept of data over time, enabling drift detection in machine learning models.
  • Category
    Marketing Automation
  • Subcategory
    Leads
  • Date Added
    2016-05-09

Specification

  • Metric Type
    general
  • Dimensional
    No
  • Decimal Digits
    No
  • Currency Units
    No
  • Granularities
    Daily, weekly, monthly, quarterly, yearly and all-time.
  • Custom Relative Periods
    Yes
  • Data Availability
    At sync, it ranges from 3 years ago to now.
  • Retroactive Data Updates
    Newly synced data is merged with existing data, replacing values for matching periods.
  • Future Data Available
    No

Visualization

  • Cumulative Graph
    Yes
  • Favorable Trend
    increasing
  • Media Creatives
    No
  • Forecasts
    Yes
  • Benchmarks
    Yes

Track all of your key business metrics from one screen

GET STARTED
Databox app preview