مدلسازی سیستم مدیریت دانش با نمایه سازی ماشینی مبتنی بر اصطلاحنامه در یک شرکت فعال در تکنولوژی اطلاعات Modeling a Knowledge Management System through Thesaurus-Based Automatic Indexing in an IT Company (PHD Thesis)

Abstract

The aim of this study is modeling a Knowledge Management System through Thesaurus-Based Automatic Indexing in an IT Company.
The purpose of this research is evaluating the current status of knowledge management in IT companies and exploring the problems as a result to seek the possible solutions to improve the KM implementation. The results of this research can be utilized in self-improvement of knowledge management in IIN, and also applied as a reference for other IT enterprises when they are trying to deploy knowledge management solution for the organization, so that potential obstacles can be avoided, whereas risks can be reduced as well. Upon the importance of knowledge management on IT company’s competitive edge in competing business environment, this research could be regarded as a resolution to address problem if it exists, whereas achieve competitive advantages.
At the end a construction products’ database, originally classified according to a hierarchical structure, was used in this analysis. The results could demonstrate the effectiveness and applicability of automated document classification methods for management information systems.
One limitation of the existing inter-organizational information systems is the reliance on manual classification methods conducted by human experts. With the growth in the use of information technologies by construction companies, specially IIN, the increasing availability of electronic documents, and the development of model-based systems, manual classification becomes impractical.
This project presents a way to improve information organization and access in inter-organizational construction management systems based on methods for automated hierarchical classification of construction documents according to CICSs items. In order to accomplish this goal, a combination of techniques from the areas of information retrieval and text mining was explored.
The use of thesaurus-based indexing is a common approach for increasing the performance of document retrieval. With the growing amount of documents available, manual indexing is not a feasible option. Statistical methods for automated document indexing are an attractive alternative. I argue that the quality of the thesaurus used as a basis for indexing in regard to its ability to adequately cover the contents to be indexed is of crucial importance in automatic indexing because there is no human in the loop that can spot and avoid indexing errors. I propose a method for thesaurus evaluation that is based on a combination of statistical measures and appropriate visualization techniques that supports the detection of potential problems in a thesaurus. I describe this method and show its application in the context of two automatic indexing tasks. The examples show that the methods indeed eases the detection and correction of errors leading to a better indexing result
This research used case study materials for a single company were assembled in order to raise some of the primary issues and to provide the ability to drill down on a common example.
Results show that it does not require the manual assignment of metadata (keywords or index terms) to all documents in the information system. Manual assignment of metadata is a tedious task. It is also hard to achieve consistency when a large number of people from different publishers are adding documents to the system.
It does not need the utilization of a controlled vocabulary that would only be effective if it was accepted as a standard by a publisher and adopted by all publishers of other management information system.
It uses already existing standards (LC,ISI and ULRICH) to define the categories that will be used for classification;
and it facilitates the creation of automated mapping mechanisms of document subjects.
systems, manual classification becomes impractical. One example of the limitations of manual classification is the time and effort that would be required to classify all documents created in a construction project (contracts, specifications, meeting minutes, change orders, field reports, and requests for information, among others), according to all components of a CICS1.
A second problem that exists in available systems is the lack of support for differences in vocabularies and naming conventions. This problem can be illustrated by the case in which an architect gives a name for a particular object in a project model. Since there is usually no standard vocabulary systems of publishers, The previously mentioned limitations and the push towards fully integrated and automated system justify the need for the development of automated classification methods for construction classification system that can adapt to different classification frameworks.