In any larger engineering setting, there is a huge number of documents that engineers and others need to use and be aware of in their daily work. To improve the handling of this amount of documents, we propose to view it under the angle of a new domain for professional search, thus incorporating search engine knowledge into the process. We examine the use of Information Retrieval (IR), Recommender Systems (RecSys), and Knowledge Management (KM) methods in the engineering domain of Knowledge-based Engineering (KBE). The KBE goal is to capture and reuse knowledge in product and process engineering with a systematic method. Based on previous work in professional search and enterprise search, we explore a combination of methods and aim to identify key issues in their application to KBE. We list detected challenges, discuss information needs and search tasks, then focus on issues to solve for a successful integration of the IR and KBE domain and give a system overview of our approach to build a search and recommendation tool to improve the daily informationseeking workflow of engineers in knowledge-intense disciplines. Our work contributes to bridging the gap between Information Retrieval and engineering support systems.
Knowledge and information resources play a pivotal role in enterprises and are valuable for solution reuse and learning through information access. However, identifying relevant information from a rapidly growing number of unstructured resources is challenging for users. We discuss a personalized information access tool for professional workplaces based on the recommender systems to provide relevant documents for users in specific work contexts based on domain-specific ontologies. Our use case is a multidisciplinary engineering project building an energy-efficient vehicle. We provide an indepth analysis of document corpus characteristics of this real-life shared engineering workspace to understand the content and context of documents using information retrieval methods and semantic annotations. Upon this, we build a contextual ontology as our knowledge domain for the recommender system. We validate our ontology-based content matching approach by evaluating the level of retrievability and coverage of the ontology against the indexed document corpus through experiments on the corpus and ontology. Our results provide insight into engineers' document workspaces and show that even a simple domain ontology is able to match a majority of documents from a domain-oriented corpus. The findings support our approach of using ontology-based recommendation for domain-specific workspaces.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.