Information Organization and Storage

Organizing and storing information may be facilitated with the application of information technology. Traditional data processing technologies were first used to raise work efficiency, whether on the office floor or the shop floor. The operational use of computers generated an abundance of detailed information about transactions, customers, service calls, resource utilization, and so on.

While such systems are tuned to provide high through put performance, they are inefficient at and sometimes incapable of retrieving the information that decision makers need to have for planning and problem solving.

Organizations with significant volumes of transactional information could need to reorganize and unify operational data from several sources, and provide friendly but powerful analysis tools that allow decision makers to trawl the raw data for strategic insight, so that, for example, they can discover patterns and opportunities buried in the lodes of data about customer transactions or service calls. The information assets of an organization are not confined to the transactional; they vary from the highly ordered to the ephemeral, and some of the most valuable information may be hiding in sales reports, office memos, study reports, project documents, photographs, audio recordings, and so on.

The organization, storage, and retrieval of textual and unstructured information will become a critical component of information management. The learning organization needs to be able to find the specific information that best answer a query, and to collate information that describes the current state and recent history of the organization.

Well integrated archival policies and records management systems will enable the organization to create and preserve its corporate memory and learn from its history.

The potentially severe consequences of the loss or inability to find vital documents are driving organizations to seek more versatile information storage and retrieval systems that can capture, store and retrieve text and other unstructured data. Instead of efficiency, the overall system requirement now becomes flexibility—the system should capture hard and soft information, support multiple user views of the data, link together items that are functionally or logically related, permit users to harvest the knowledge that is buried in these resources, and so on. Because the same information can be relevant to a range of different problem situations, it becomes necessary to represent and index the unstructured information by several methods.

The development of automated indexing systems makes it increasingly feasible to adopt a user-centered approach to indexing, over and above document-oriented indexing that represents the document’s content. In the user-centered approach, indexing can be done on two levels: the first reflects topic and other predetermined features; the second is tailored to situational requirements such as the level of treatment, whether general or specific (Fidel 1994). User-centered indexing may also be request-oriented, in which case the index language is built from an analysis of user requirements and is then used as a checklist to index documents (Soergel 1985).

The underlying idea is to anticipate user requests and check each document when it is being indexed against a list of anticipated requests. A combination of document-oriented and user-oriented indexing approaches has the potential to significantly improve information retrieval performance as well as user satisfaction. Given the amount of textual material in any organization, text information management will become as important as database management, and text retrieval applications will one day be as commonplace as word processing or spreadsheets. Today’s text retrieval engines and development tools have attained new levels of functionality and versatility.

Some of the newer systems make use of semantic networks of word meanings and links derived from dictionaries and thesauri to allow users to search by querying in natural language, choosing concepts and specific word meanings, and controlling the closeness of match.

Other systems simplify the development of text retrieval applications across heterogeneous database environments using a common access and programming interface based on industry standards. http://www.InfosDemocracy.com