سیستم توصیه گر مبتنی بر گردش کار

This paper proposes a workflow-based recommender system model on supplying proper knowledge to proper members in collaborative team contexts rather than daily life scenarios, e.g., recommending commodities, films, news, etc. Within collaborative team contexts, more information could be utilized by recommender systems than ordinary daily life contexts. The workflow in collaborative team contains information about relationships among members, roles and tasks, which could be combined with collaborative filtering to obtain members' demands for knowledge. In addition, the work schedule information contained in the workflow could also be employed to determine the proper volume of knowledge that should be recommended to each member. In this paper, we investigate the mechanism of the workflow-based recommender system, and conduct a series of experiments referring to several realworld collaborative teams to validate the effectiveness and efficiency of the proposed methods.

This study concerns knowledge recommender systems for collaborative team contexts, rather than general situations in daily life, e.g., recommending commodities, news, films to customers. Among a collaborative team, members usually come from diverse disciplines, each with particular expertise and contribution from their relevant areas. Thus, their demands for knowledge are also different from each other. Recommender system provides a platform to deliver right knowledge in the right context to the right person in the right volume [27,34,36]. This paper proposes a workflow-based recommender system model, which is oriented to the collaborative team environment. Within this context, more information could be utilized by recommender systems, comparing to ordinary daily life situations. Work- flow is one type of collaborative processes and it virtually exists behind every collaborative team [37,38]. The workflow in the collaborative team environment contains members-roles-tasks reference information that describes which member plays which roles or fulfills which tasks. This reference information could be combined with collaborative filtering to obtain members' demands for knowledge. It ensures that knowledge resources in proper domains will be recommended to proper members in collaborative team. Moreover, the volume of those recommended knowledge resources should also be proper for each member. Otherwise, too much knowledge is recommended to some busy members, which will cause informationoverload and interruption to them. In our study, the work schedule information contained in the workflow is utilized to determine the proper volume of recommended knowledge for each member. This paper investigates the mechanism of the workflow-based recommender system, and conducts a series of experiments referring to several real-world collaborative teams so as to validate the effectiveness and efficiency of the proposed methods. The rest of this paper is organized as follows. Some related works done by other scholars are briefly introduced in the next section. In Section 3, we introduce the application background: collaborative environment, which is the basis for our proposed method. Then, Section 4 addresses the general framework of the workflow-based recommender system, and analyzes two key technical issues. Sections 5 and 6 investigate those issues in detail respectively: workflow-based collaborative filtering, and recommendation volume control by using the schedule information in workflow. For performances evaluation, several experiments are conducted to validate the proposed model and methods in Section 7. Closing remark and summary are then outlined in the last section.

This paper introduces a workflow-based recommender system model for collaborative team environment. Two workflow-centric approaches for mining team members' knowledge demands and determining proper recommendation volume are proposed. This study paves the way for implementing a platform which would ensure that a proper volume of proper knowledge resources could be recommended to the proper members among the collaborative team. However, there exist some limitations for the current model and methods, which need further studies in future: (1) The proposed methods consider expert–novice influence on members' demands for knowledge. However, the current study has not mentioned how ‘expert’ a new member is viewed. In Section 4.3, the manner in which hl is used has not taken into account how much expertise an individual brings into the collaborative team. This is actually a new user cold starting problem. In future studies, the agency theory could be applied to how knowledge workers address their knowledge needs. In this way, it may improve on or replace the current methods of using hl. (2) In experiments, we are not able to use all possible combinations of parameters. Currently, those settings are determined according to experience. As to different collaborative teams, the ‘optimal’ settings are actually different from each other. There is no universal setting that could adapt to all contexts with the best performance. The sensitivity analysis of those parameter settings for different application environments should be conducted in future studies.