Browsing by Author "Chang, Pei-Cheng"

Now showing 1 - 4 of 4

An exploratory study of incorporating task-stage identification into a long-term document support process
(2006)
Wu, I-Chin
;
Chang, Pei-Cheng
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
With the increasing amount of information (and knowledge) in organizational memories and the phenomenal growth of the World Wide Web, knowledge workers are increasingly suffering from information overload Because of the growing complexity of knowledge-intensive tasks, a profiling technique is needed to deliver task-relevant documents to workers. In this study, we propose a task-stage identification technique that provides effective document support throughout the execution of a task. A pattern similarity analysis method based on the correlation value is used to identify a worker\'s task-stage. The results are incorporated into a profile adaptation process to generate the worker\'s current task profile. We conducted a pilot study in a research institute. The results confirm the effectiveness of the proposed document support process in a knowledge-intensive working environment.
Learning Dynamic Information Needs: A Collaborative Topic Variation Inspection Approach
(2009-12-01)
Wu, I-Chin
;
Liu, Duen-Ren
;
Chang, Pei-Cheng
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
For projects in knowledge-intensive domains, it is crucially important that knowledge management systems are able to track and infer workers' up-to-date information needs so that task-relevant information can be delivered in a timely manner. To put a worker's dynamic information needs into perspective, we propose a topic variation inspection model to facilitate the application of an implicit relevance feedback (IRF) algorithm and collaborative filtering in user modeling. The model analyzes variations in a worker's task-needs for a topic (i.e., personal topic needs) over time, monitors changes in the topics of collaborative actors, and then adjusts the worker's profile accordingly. We conducted a number of experiments to evaluate the efficacy of the model in terms of precision, recall, and F-measure. The results suggest that the proposed collaborative topic variation inspection approach can substantially improve the performance of a basic profiling method adapted from the classical RF algorithm. It can also improve the accuracy of other methods when a worker's information needs are vague or evolving, i.e., when there is a high degree of variation in the worker's topic-needs. Our findings have implications for the design of an effective collaborative information filtering and retrieval model, which is crucial for reusing an organization's knowledge assets effectively.
Measuring the variation in task-needs for knowledge delivery: A profiling via collaboration technique
(2007)
Liu, Duen-Ren
;
Wu, I-Chin
;
Chang, Pei-Cheng
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
Effective knowledge. management (KM) in a knowledge-intensive working environment requires an understanding of workers' information needs for tasks, (task-needs), so that they can be provided with appropriate codified knowledge (textual documents) when performing long-term tasks. This work proposes a novel profiling technique based on implicit relevance feedback and collaborative filtering techniques that model workers' task-needs. The proposed profiling method analyses variations in workers' task-needs for topics (i.e., topic needs) in a topic taxonomy over time. Variations in the topic needs of similar workers' are used to predict variations in a target worker's topic needs and adjust his/her task profile accordingly. Experiment results suggest that considering variations In the topic needs of similar workers' during the profile adaptation process Is effective in improving the precision of document retrieval.
Toward incorporating a task-stage identification technique into the long-term document support process
(2008-09-01)
Wu, I-Chin
;
Liu, Duen-Ren
;
Chang, Pei-Cheng
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
Effective knowledge management in a knowledge-intensive environment can place heavy demands on the information filtering (IF) strategies used to model workers' long-term task-needs. Because of the growing complexity of knowledge-intensive work tasks, a profiling technique is needed to deliver task-relevant documents to workers. In this study, we propose an IF technique with task-stage identification that provides effective codification-based support throughout the execution of a task. Task-needs pattern similarity analysis based on a correlation value is used to identify a worker's task-stage (the pre-focus, focus formulation, or post-focus task-stage). The identified task-stage is then incorporated into a profile adaptation process to generate the worker's current task profile. The results of a pilot study conducted in a research institute confirm that there is a low or negative correlation between search sessions and transactions in the pre-focus task-stage, whereas there is at least a moderate correlation between search sessions/transactions in the post-focus stage. Compared with the traditional IF technique, the proposed IF technique with task-stage identification achieves, on average, a 19.49%, improvement in task-relevant document support. The results confirm the effectiveness of the proposed method for knowledge-intensive work tasks. (c) 2007 Elsevier Ltd. All rights reserved.