بهبود مستمر سیستم های مدیریت دانش با استفاده از متدولوژی شش سیگما
|کد مقاله||سال انتشار||تعداد صفحات مقاله انگلیسی||ترجمه فارسی|
|6838||2013||9 صفحه PDF||سفارش دهید|
نسخه انگلیسی مقاله همین الان قابل دانلود است.
هزینه ترجمه مقاله بر اساس تعداد کلمات مقاله انگلیسی محاسبه می شود.
این مقاله تقریباً شامل 5500 کلمه می باشد.
هزینه ترجمه مقاله توسط مترجمان با تجربه، طبق جدول زیر محاسبه می شود:
- تولید محتوا با مقالات ISI برای سایت یا وبلاگ شما
- تولید محتوا با مقالات ISI برای کتاب شما
- تولید محتوا با مقالات ISI برای نشریه یا رسانه شما
پیشنهاد می کنیم کیفیت محتوای سایت خود را با استفاده از منابع علمی، افزایش دهید.
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Robotics and Computer-Integrated Manufacturing, Volume 29, Issue 3, June 2013, Pages 95–103
Knowledge retrieval is a decisive part of the performance of a knowledge management system. In order to enhance retrieval accuracy, an effective performance evaluation mechanism is necessary. Nowadays, there is not a standard evaluation framework for knowledge retrieval evaluation, because the evaluation set up is still technology-dependent, focusing on specific elements of the search context. The laboratory-based evaluation is not suitable to evaluate the knowledge retrieval process, since knowledge is dynamic, constantly changing and evolving. Besides, ambiguous query is also an important factor for the performance of knowledge retrieval systems. In order to improve the performance of knowledge retrieval, this paper proposes an evaluation mechanism using Six Sigma methodology to help developers continuously control the knowledge retrieval process. Specifically, this study involves the following tasks: (i) proposes a general knowledge retrieval framework based on the analysis result of knowledge retrieval, (ii) designs the knowledge retrieval evaluation framework using Six Sigma's Define-Measure-Analyze-Improve-Control (DMAIC) process and (iii) develops the related technologies to implement the knowledge retrieval evaluation mechanism. The knowledge retrieval evaluation mechanism allows system developers to maintain the knowledge retrieval system with ease and meanwhile enhance the accuracy.
Knowledge has been recognized as the key resources of business survival and success in knowledge economy. Knowledge, while made up of data and information, can be thought of as much greater understanding of a situation, relationships, causal phenomena, and the theories and rules (both explicit and implicit) that underlie a given domain or problem . Easier access to data and documents can help firms reduce the development cycles and lead times . Therefore, knowledge management which consists of Create, Storage, Retrieval, Transfer and Reuse of knowledge has become an important approach to improve the competitive advantage of enterprises . To transmit the right knowledge to the right people at the right time, knowledge retrieval is the major part of knowledge management. However, knowledge retrieval is a time-consuming task in the large knowledge base. It is impossible for users to filter all knowledge to determine the needed knowledge in large-scale knowledge bases, which involves a large amount of knowledge , and it is difficult for users to decide what knowledge is needed before they know it. Knowledge Retrieval Systems (KRS) which is the useful retrieval systems for supporting knowledge discovery, organization, storage and retrieval, guarantees access to large corpora of unstructured data . In order to enhance retrieval accuracy, performance evaluation is also an important task for knowledge retrieval systems. Nowadays, there is not a standard evaluation framework for knowledge retrieval evaluation, because the evaluation set up is still technology-dependent, focusing on specific elements of the search context . Knowledge retrieval evaluation usually refers to the methods of information retrieval evaluation. The most commonly used methodology of information retrieval systems evaluation needs a test collection, which contains a document collection, a set of topical queries and a set of relevance assessments identifying the documents that are topically relevant to each query . However, knowledge can be represented in different ways and stored in different types. Knowledge is dynamic, constantly changing and evolving . Knowledge repository is under continuous growth in the real world. Hence, real search are often out of the evaluation scope. Besides, ambiguous query usually results in the lower accuracy of knowledge retrieval. Variance between queries is larger than the variance between systems . Therefore, user queries processing is also an important factor influencing the performance of knowledge retrieval. Six Sigma, a customer-focused and data-driven quality strategy, is a rigorous and systematic methodology that utilizes collected information and statistical analysis to measure and improve performance. The philosophy of Six Sigma is to keep a process within its limits so almost no defects occur . In order to improve the performance of knowledge retrieval, this paper proposes a knowledge retrieval evaluation mechanism using Six Sigma to help developers to continuously control the knowledge retrieval process. To achieve this objective, this paper first proposes a general knowledge retrieval framework based on the analysis result of knowledge retrieval and then designs the knowledge retrieval evaluation framework using Six Sigma's Define-Measure-Analyze-Improve-Control (DMAIC) process to continuously improve the efficiency of knowledge retrieval. Finally, this paper develops the related technologies to implement the knowledge retrieval evaluation mechanism. The evaluation mechanism is an effective tool to facilitate the knowledge retrieval process more robust and, hence, ensure satisfactory performance in support of problem solving, decision making, and knowledge innovation.
نتیجه گیری انگلیسی
This paper proposes a knowledge retrieval evaluation mechanism which consists of the performance measuring, monitoring and diagnosis to continuously control the quality of the knowledge retrieval process. In order to realize this mechanism, a performance measure, FAveP, is designed, which integrates the F-mean and the Average Precision. Following that, a rule-based reasoning engine is designed that identifies the causes of errors to adjust the knowledge retrieval system accordingly. Finally, an adaptive p-chart is proposed to monitor the performance of the knowledge retrieval system. The proposed mechanism allows system developers to maintain the knowledge retrieval system more easily. If the defect rate is over the upper control limit, the evaluation mechanism will trigger the revise function to adjust the knowledge retrieval system to improve the accuracy of knowledge retrieval. Therefore, the knowledge retrieval system will deliver higher accuracy to respond to knowledge requirements, and users will have higher intention to use it. Consequently, knowledge can be transmitted and re-used efficiently through the knowledge retrieval system, which will lead to improved new knowledge creation process. Furthermore, user's perspective has been recognized as an important factor for knowledge retrieval. A specific user's search context has a significant impact on performance evaluation of knowledge retrieval systems. Hence, knowledge retrieval evaluation needs to be explored via more dimensions in the future.