To read the full version of this content please select one of the options below:

An approach to task-oriented knowledge recommendation based on multi-granularity fuzzy linguistic method

Ming Li (School of Business Administration, China University of Petroleum, Beijing, China)
Mengyue Yuan (School of Business Administration, China University of Petroleum, Beijing, China)
Yingcheng Xu (China National Institute of Standardization, Beijing, China)

Kybernetes

ISSN: 0368-492X

Article publication date: 2 March 2015

Abstract

Purpose

In organizations, knowledge intensive activities are mainly task oriented. Finding relevant completed tasks to the new task and providing task-related knowledge to workers facilitate the knowledge reuse. However, relevant tasks are not easily found in the huge amount of completed tasks. The purpose of this paper is to assist the worker to find the required knowledge for the task at hand by reusing the knowledge related to relevant competed tasks.

Design/methodology/approach

First, the task profile is constructed. Relevant degrees to categories which tasks to are derived by multi-granularity fuzzy linguistic method. The stages of completed tasks are identified by the modified KNN method. Second, similar completed tasks on categories are retrieved and then the relevant tasks are selected from the retrieved similar tasks by multi-granularity fuzzy linguistic method. Third, the worker’s current task stage is derived by calculating the similarity between the rated knowledge and the knowledge in the stage of completed tasks. Finally, the knowledge is recommend based on stage relevance, relevance of the completed tasks and importance of the knowledge.

Findings

The proposed method helps the worker find the knowledge related to the task at hand by finding and reusing the completed tasks. The experimental results show that the proposed method performs well and can fulfill the worker’s’ knowledge needs. The use of the linguistic term set with preferred granularities instead of precise numbers facilitates the expression of the opinions. The recommendation stage by stage makes the knowledge recommended more precisely. The obtained linguistic weight of the knowledge makes the recommended results understood more easily than the numerical values.

Research limitations/implications

In the study, the authors just focus on the codified knowledge recommendation. However, there is another kind of knowledge named tacit knowledge, which exists in the mind of the experts. The constructing and updating of the expert profile can be investigated. Meanwhile, the new recommendation method which considers more factors also needs to be studied further.

Practical implications

The paper includes implications for the development of the knowledge management system. The proposed approach can be applied as a tool of knowledge sharing. It facilitates the finding of the knowledge that is related to the task at hand.

Originality/value

The paper provides new ways to find the relevant tasks and the related knowledge to the task at hand. Meanwhile, the new method to recommend the knowledge stage by stage is also proposed. It expands the research in the knowledge sharing and knowledge recommendation.

Keywords

Acknowledgements

The research is supported by the National Natural Science Foundation of China under Grant No. 71101153, 71301152.

Citation

Li, M., Yuan, M. and Xu, Y. (2015), "An approach to task-oriented knowledge recommendation based on multi-granularity fuzzy linguistic method", Kybernetes, Vol. 44 No. 3, pp. 460-474. https://doi.org/10.1108/K-10-2014-0207

Publisher

:

Emerald Group Publishing Limited

Copyright © 2015, Emerald Group Publishing Limited