Browse Prior Art Database

A system and method for NL interactions over heterogeneous data from real time bound systems Disclosure Number: IPCOM000240410D
Publication Date: 2015-Jan-29
Document File: 2 page(s) / 73K

Publishing Venue

The Prior Art Database


Natural language queries have been in an active research topic since 1960's. But providing real time cognitive insights for end user queries need techniques for optimizing the transformed query by identifying valid rewriting rules of expressions . This also mandates conforming feasibility of support by leveraging schematic insights of underlying systems. This is required as enterprise infrastructure and physical assets contain rich structured contextual information about their status, contents, and that of users, apps, tasks. Also retrieval of contextual information is must for enriching search accuracy with cognitive systems. In this article, we propose a natural Language based centralized end user portal system that augments IT domain focusing on information access, query, reasoning that enables user driven root cause analysis that span multiple collaborating systems through semantically and structurally inter-related entities. Enabling free text interactions with multiple systems that are bound runtime and contextual aggregation of response from heterogeneous mix of structured and unstructured data using nested ontological models is also proposed.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 100% of the total text.

Page 01 of 2

A system and method for NL interactions over heterogeneous data from real time bound systems

We propose a Logical framework to represent natural language questions in a form that resolve the targeted systems from the registered list of systems and subsequently validates the feasibility of answering the question at least partially. The framework basically comprise the following :
1.An NLP pipeline that facilitates System and technology specific read/write queries 2. An extensible nested set of ontology models derived from a standard schema to glue together what user types, what she means in system terms, and how we retrieve information from underlying systems 3. Filtering and aggregation of retrieved information from multiple sources towards compiling a response or augmenting WEA with a refined query for improved accuracy. The outline of the frame work is depicted in Fig. A.

Fig A. Architecture for Cognitive Insights over Structured Data.

As a motivating example, we have considered accessing ticket data from SCCD ( An IBM Tivoli Product for IT management) . The following figure Fig. B, illustrates the run time interactions of the different components for accessing ticket data based on end user queries..


Page 02 of 2

Fig B. Run time Interactions for Accessing Ticket Data