Extraction and processing of procedural experiential knowledge in workflows - quality, interactivity, and transferability

点击翻译按钮获取中文摘要

The goal of this project is to investigate new methods in Case-Based Reasoning and related fields for extracting, representing and processing procedural experiential knowledge in Internet communities. Procedural knowledge is the knowledge exercised in the performance of some task. Similar to a plan, it describes how to do a certain thing or to achieve a certain goal through a sequence of steps. In Internet forums numerous records of instances of such procedural experiential knowledge is found. In the work performed during the first funding period of this project, we focus on workflow representations of procedural experiential knowledge. Our main research addressed the extraction of workflows from textual sources in Internet Communities, the similarity-based retrieval of workflows for a particular goal of a user, and the automatic adaptation of retrieved workflows. With this proposal, we apply for a second funding period to extend our work in four respects:Adaptation Quality: While in our previous research, we developed several methods that enable the automatic adaptation of workflows by using adaptation knowledge automatically learned from workflow repositories, the quality of the adapted workflows is difficult to control. Therefore, we aim at investigating new methods for assessing the quality of automatically adapted workflows as well as methods to assess the impact of the learned adaptation knowledge on the workflow quality.Interactivity: The workflow retrieval and adaptation methods developed so far are fully automatic and assume a fully developed query, which is difficult for a user to specify upfront. Therefore, we aim at developing new methods for conversational CBR that enable fully interactive problem solving involving retrieval and adaptation of workflows.Transfer Learning: The adaptation methods investigated so far require existing procedural knowledge of significant volume in order to learn enough adaptation knowledge. This makes it difficult to address small or newly emerging domains in which procedural knowledge is still sparse. Therefore, we aim at investigating whether transfer learning methods can be used to improve learning of adaptation knowledge by transferring knowledge from a different, but related domain with substantial procedural knowledge.Exploring New Application Domains: So far, we demonstrated our methods primarily in the domain of cooking workflows. In the second funding period, we aim at broadening the experimental basis for the whole project by exploring workflow and business process model repositories available in existing repository collections that become (recently) available within business process research. We will select several repositories and transform them semi-automatically into a case bases of semantic workflows or process models to be used as reference data set for the planned experimental work.