Architecture and method for providing insights in networks domain

Inventors

Duggimpudi, Maria Bala • Moursy, Abdelhamid • Ali, Elshaimaa • Raghavan, Vijay V.

Assignees

University of Louisiana at Lafayette

Abstract

The invention is ontology-based architecture for performing semantic data mining for insights. The major components of the architecture are illustrated using Network Ontology (NO), an ontology generated for the network domain for delivering improved network management. In addition, algorithms for summarizing performance profiles in the form of rank tables and for extracting insight rules (concrete action plan) from the rank tables are presented. By using this approach, an actionable plan for assisting decision maker can be obtained, as domain knowledge is incorporated in the system. Experimental results on a network dataset show that this model provides an optimal action plan for a network to improve its performance by encoding data-driven rules into the ontology and suggesting changes to its current network configuration.

Core Innovation

The invention provides an ontology-based insight-as-a-service for network management by building a Network Ontology for a network domain. It builds the wireless network ontology by manually extracting semantics from a relational database and representing the resulting knowledge using RDF triples stored in a triple-store and reasoner. The ontology is then used as the basis for generating and encoding insight rules for a decision maker.

The method applies a rank table generation algorithm to discretize controllable and uncontrollable parameters and performance metrics into categories, and to generate rank tables by aggregating performance metrics into rank values. Rank tables are used as inputs to an insight rule generation algorithm. The insight rules recommend changes to controllable parameters by using distances to combinations in a target healthy rank category and by selecting an optimal row based on the minimum distance criteria.

An architecture supports operation by a graphic user interface that provides experiments lookup and selection. A selected experiment is submitted, and the system invokes a reasoner over the encoded insight rules to infer recommended configurable settings. The inferred recommendations are displayed to the decision maker, where each network configuration responds to a different experiment.

Claims Coverage

The independent claim is clm-00001, which defines the end-to-end workflow for providing insights using a wireless network ontology, rank tables, insight rules, a GUI-driven experiment lookup, and a reasoner that displays inferred recommended configurable settings. The main claim coverage includes a method that builds the ontology from a relational database, generates rank tables, generates and encodes insight rules, runs experiment selection, invokes reasoning, and displays at least one inferred recommended configurable setting.

Overall, the claim coverage centers on an ontology-based insight workflow: a manually extracted wireless network ontology is combined with rank table generation, insight rule generation, and ontology encoding; a GUI selects an experiment, the reasoner is invoked, and at least one inferred recommended configurable setting is displayed, with each configuration tied to a different experiment.

Stated Advantages

Documented Applications