International Business Machines Corporation

Theodor Razvan Ionescu

Ruchi Mahindru

Yu Deng

Harigovind Venkatraj Ramasamy

Soumitra Sarkar

Long Wang

Simon J Kofkin-Hansen

Christopher Moss

Richard Christopher Smith

A mechanism is provided for identifying patterns of a set of software applications instances from their documents. The computer-implemented method begins with constructing different attribute vector types using a knowledge ontology. The knowledge ontology captures semantics based on keywords associated with resource attributes derived from one or more documents related to at least a portion of these software application instances. A knowledge base is built from the attribute vector types and the documents of these application instances. These are merged into the knowledge base with the knowledge base previously built from previous software application instances. Analytics are performed on the knowledge base to identify at least one of common patterns of deployments, configurations, or other attribute vector types, or a combination thereof.

Identifying patterns of a set of software applications

Growing community of inventors

Bucharest, Romania

Theodor Razvan Ionescu