Browse Prior Art Database

System and Method for Optimizing Search on Enterprise Wikis

IP.com Disclosure Number: IPCOM000224400D
Publication Date: 2012-Dec-18

Publishing Venue

The IP.com Prior Art Database

Abstract

A hierarchical organizational structure gives a large business enterprise the ability to segregate large sections of the company’s business into semiautonomous groups, called domains. A focussed and efficient enterprise search system for a particular domain, needs to identify the knowledge repositories of interest to the domain, and index only those in its system. Enterprise wikis hold a significant amount of unstructured business data, and are therefore important enterprise knowledge repositories. In this paper, we first outline the limitations of conventional web content ingestion techniques in handling the web of enterprise wikis. We then present how, high-precision, high-recall, domain-specific search systems can be developed over this web of enterprise wikis.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 12% of the total text.

Page 01 of 12

System and Method for Optimizing Search on Enterprise Wikis

Abstract. A hierarchical organizational structure gives a large business enterprise the ability to segregate large sections of the company's business into semi-autonomous groups, called domains. A focused and efficient enterprise search system for a particular domain needs to identify the knowledge repositories of interest to the domain, and index only those in its system. Enterprise wikis hold a significant amount of unstructured business data, and are therefore important enterprise knowledge repositories. In this paper, we first outline the limitations of conventional web content ingestion techniques in handling the web of enterprise wikis. We then present how, high-precision, high-recall, domain-specific search systems can be developed over this web of enterprise wikis.

1 Introduction

Ever-growing popularity of Web 2.0 applications, like wikis, forums, blogs, content management systems, etc, has resulted in tremendous growth in unstructured business data. These applications focus on the ability for people to collaborate and share information online, and, have essentially migrated data from the user's personal desktop to the enterprise's managed, shared data stores. With organizations becoming more complex, it is no wonder that a large percentage of a consultant's time is wasted on searching these data stores or recreating existing content. Focused and efficient enterprise search solutions can help in substantially reducing this time, and thereby, help in increasing productivity. Before delving deep into building focused enterprise search solutions, it is important to understand the general organizational structure of enterprises.

A hierarchical organizational structure gives a large business enterprise, the ability to segregate large sections of the company's business into semi-autonomous groups. Teams in these semi-autonomous groups work independently, yet, very frequently, need to align with each other, e.g., teams responsible for proposing new business solutions would often want to interact with the solution delivery teams, to understand, the risks involved, the assets already available from past engagements, etc. We refer to these semi-autonomous groups as enterprise domains, in tune with Albertas et. al. [3], who defined domain engineering as a discipline that aims to support systematic reuse of business solutions and supporting software, focusing on modeling common knowledge in a problem domain. Furthermore, we make the

1


Page 02 of 12

assumption that - each enterprise domain can be modeled as a set of queries on a finite set of topics - and therefore, any focused and efficient enterprise search solution for a domain should be tuned around the corresponding topical domain queries.

Wiki is an important Web 2.0 application, especially in an enterprise framework. Basically, an enterprise wiki is a mini web, consisting of titled, structured or unstructured web page...