iLCM - A virtual research infrastructure for large-scale qualitative data

ILCM项目将开发一个综合研究环境，用于分析软件即服务架构(SaaS)中的结构化和非结构化数据。该项目涉及利用文本挖掘对定性数据进行定量研究的要求，以及对社会科学中以数据为导向的研究设计的可再现性的要求。研究环境iLCM基于莱比锡语料库挖掘器(LCM)，这是一个分散的SaaS应用程序，专为分析大量报纸文本而设计，是为BMBF资助的政治学电子人文项目开发的。为了将LCM-Prototype用于通用研究问题，它将被扩展为具有新功能。此外，LCM的通用文本挖掘工具将得到开放式研究计算环境(ORC)的补充，该环境适用于称为笔记本的活动和可执行文件。这种OCR环境允许以灵活的方式将LCM可以从文本中提取的语义结构与其他数据相链接。这允许开发从项目特定要求派生的单独研究设计。分析工作流可以被组织和存储为笔记本，即其口头描述的脚本，并且可以作为活动的和可执行的文档与研究数据一起发布。作为存档研究数据的服务提供者，地理信息系统将在该项目范围内为笔记本的执行、出版和存档开发一项中央服务。笔记本以及原始和中间数据的出版将使研究结果和方法可复制、可共享和可重复使用。提出者期待着iLCM在研究和教学方面在新兴计算社会科学领域的发展上向前迈出一大步。

The project iLCM will develop an integrated research environment for the analysis of structured and unstructured data in a Software as a Service-architecture (SaaS). The project addresses the requirements of quantitative research in qualitative data with text mining and the requirements on the reproducibility of data-driven research designs in the social sciences. The research environment iLCM is based on the Leipzig Corpus Miner (LCM), a decentralized SaaS-application designed for the analysis of large amounts of newspaper texts which has been developed for a BMBF funded eHumanities project in political science. To use the LCM-prototype for generic research questions it will be extended with new functionalities. Additionally, the generic text mining tools of the LCM will be supplemented with an Open Research Computing-environment (ORC) for active and executable documents referred to as notebooks. This OCR-evironment allows to link the semantic structures which can be extracted from texts by the LCM with other data in a flexible way. This allows to develop individual research designs that derive from project specific requirements. Analysis workflows can be organized and stored as Notebooks, i.e. scripts of their verbal descriptions, and can be published in company with the research data as active and executable documents. GESIS as a service provider for archiving research data will develop a central service for the execution, publishing and the archiving of notebooks within the scope of this project. The publishing of notebooks along with source- and intermediate data will make research results and methodologies reproducible, shareable and reusable. The proposers expect a large step forward in the development of the emerging field of Computational Social Science from iLCM in research and teaching.