Machine learning-based interactive visual monitoring tool for high dimensional data sets across multiple KPIs
Described are computing systems and methods configured to detect a small, but meaningful, anomaly within one or more metrics associated with a platform. The system displays visuals of the metrics so that a user monitoring the platform can effectively notice a problem associated with the anomaly and take appropriate action to remediate the problem. An operational visual includes a radar-based visual with a heatmap arranging metrics, and a node representing a state of the metrics. Moreover, the system uses an ensemble of unsupervised machine learning algorithms for multi-dimensional clustering of hundreds of thousands of monitored metrics. Via the visuals and the implementation of the machine learning algorithms, the described techniques provide an improved way of representing and simulating many metrics being monitored for a platform. Moreover, the techniques are configured to expose actionable and useful information associated with the platform in a manner that can be effectively interpreted.
1 . A method implemented by a computing device, the method comprising:
receiving, by the computing device, data describing a plurality of metrics being monitored for a platform;
generating, by the computing device, at least in part by using a multi-agent voting system, a visual that represents the data describing the plurality of metrics by rendering a heatmap and an object that represents a combination of the plurality of metrics, in which a movement of the object relative to the heatmap indicates a degree to which real-time data for at least one of the plurality of metrics is anomalous to observed historic data for the at least one of the plurality of metrics;
displaying, by the computing device, the visual in a user interface and receiving user input selecting one or more simulation settings to apply to the visual;
generating, by the computing device, based at least on the user input, simulated data of a simulated state of the platform for one or more of the plurality of metrics under the one or more simulation settings selected; and
updating, by the computing device, the visual based on the simulated data to represent the simulated state of the platform as an updated visual comprising the object, a second movement of the object with respect to the updated visual indicating a degree to which the simulated data for at least one of the plurality of metrics under the one or more simulation settings is anomalous to the observed historic data for the at least one of the plurality of metrics.
2 . The method of claim 1 , wherein the user input is received via an interactive slider.
3 . The method of claim 1 , wherein the visual includes an alternative visual that includes a plurality of sections and each section of the plurality of sections is associated with a performance metric, wherein the each section of the plurality of sections is dynamically colored to indicate performance with respect to the performance metric, and dynamically changes colors in real-time as the user input is received.
4 . The method of claim 1 , wherein the plurality of metrics represent performance indicators.
5 . The method of claim 4 , further comprising:
receiving, by the computing device, a user selection of the object; and
rendering, by the computing device, a web diagram that overlays the visual to visualize a representation of a relationship between a performance metric value of the object as compared with other performance metric values for the object.
6 . The method of claim 1 , wherein the visual includes a tree map visual that includes a plurality of sections and each section of the plurality of sections is associated with an attribute used to compose one or more of the plurality of metrics.
7 . The method of claim 6 , wherein at least one of a size or a color of a section of the plurality of sections indicates an amount of anomalous activity for the attribute associated with the section.
8 . The method of claim 1 , wherein the one or more simulation settings includes a traffic simulation setting.
9 . The method of claim 1 , wherein the one or more simulation settings includes a future point in time.
10 . The method of claim 1 , wherein the multi-agent voting system includes Quantile Loss Gradient Boosted Trees machine learning model-based agents.
11 . A system comprising:
a data manager module implemented at least partially in hardware of a computing device to receive data describing a plurality of metrics being monitored for a platform;
a visual manager module implemented at least partially in the hardware of the computing device to generate, at least in part via a multi-agent voting system, a visual that represents the data describing the plurality of metrics by rendering a heatmap and an object that represents a combination of the plurality of metrics, in which a movement of the object relative to the heatmap indicates a degree to which real-time data for at least one of the plurality of metrics is anomalous to observed historic data for the at least one of the plurality of metrics;
a visual rendering module implemented at least partially in the hardware of the computing device to display the visual via a user interface;
a user interface module implemented at least partially in the hardware of the computing device to receive a user input selecting one or more simulation settings to apply to the visual;
a simulation manager module implemented at least partially in the hardware of the computing device to generate, based at least in part on the user input, simulated data of a simulated state of the platform for one or more of the plurality of metrics under the one or more simulation settings selected; and
a visual updater module implemented at least partially in the hardware of the computing device to update the visual based on the simulated data to represent the simulated state of the platform as an updated visual comprising the object, a second movement of the object with respect to the updated visual indicating a degree to which the simulated data for at least one of the plurality of metrics under the one or more simulation settings is anomalous to the observed historic data for the at least one of the plurality of metrics.
12 . The system of claim 11 , wherein the visual is generated using an unsupervised artificial neural network algorithm that projects high-dimensional data onto a two-dimensional map.
13 . The system of claim 12 , wherein the unsupervised artificial neural network algorithm comprises a self-organizing map (SOM) algorithm.
14 . The system of claim 11 , wherein the user input is a recognized gesture or utterance.
15 . The system of claim 11 , wherein the one or more simulation settings includes a traffic simulation setting.
16 . The system of claim 11 , wherein the multi-agent voting system includes Quantile Loss Gradient Boosted Trees machine learning model-based agents.
17 . Computer-readable storage media comprising instructions that, when executed by one or more processing units, cause a system to perform operations comprising:
receiving data describing a plurality of metrics being monitored for a platform;
generating, at least in part by using a multi-agent voting system, a visual that represents the data describing the plurality of metrics by rendering a heatmap and an object that represents a combination of the plurality of metrics, in which a movement of the object relative to the heatmap indicates a degree to which real-time data for at least one of the plurality of metrics is anomalous to observed historic data for the at least one of the plurality of metrics;
displaying the visual in a user interface and receiving user input selecting one or more simulation settings to apply to the visual;
generating, based at least on the user input, simulated data of a simulated state of the platform for one or more of the plurality of metrics under the one or more simulation settings selected; and
updating the visual based on the simulated data to represent the simulated state of the platform as an updated visual comprising the object, a second movement of the object with respect to the updated visual indicating a degree to which the simulated data for at least one of the plurality of metrics under the one or more simulation settings is anomalous to the observed historic data for the at least one of the plurality of metrics.
18 . The computer-readable storage media of claim 17 , wherein the plurality of metrics represent performance indicators of a plurality of systems.
19 . The computer-readable storage media of claim 17 , wherein the multi-agent voting system includes Quantile Loss Gradient Boosted Trees machine learning model-based agents.
20 . The computer-readable storage media of claim 17 , wherein the visual includes a plurality of sections and each section of the plurality of sections is associated with a performance metric used to compose one or more of the plurality of metrics.