@conference {947, title = {Chatbot Explorer: Towards an understanding of knowledge bases of chatbot systems}, booktitle = {30th International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision 2022}, year = {2022}, abstract = {A chatbot can automatically process a user{\textquoteright}s request, e.g. to provide a requested information. In doing so, the user starts a conversation with the chatbot and can specify the request by further inquiry. Due to the developments in the field of NLP in recent years, algorithmic text comprehension has been significantly improved. As a result, chatbots are increasingly used by companies and other institutions for various tasks such as order processes or service requests. Knowledge bases are often used to answer users queries, but these are usually curated manually in various text files, prone to errors. Visual methods can help the expert to identify common problems in the knowledge base and can provide an overview of the chatbot system. In this paper, we present Chatbot Explorer, a system to visually assist the expert to understand, explore, and manage a knowledge base of different chatbot systems. For this purpose, we provide a tree-based visualization of the knowledge base as an overview. For a detailed analysis, the expert can use appropriate visualizations to drill down the analysis to the level of individual elements of a specific story to identify problems within the knowledge base. We support the expert with automatic detection of possible problems, which can be visually highlighted. Additionally, the expert can also change the order of the queries to optimize the conversation lengths and it is possible to add new content. To develop our solution, we have conducted an iterative design process with domain experts and performed two user evaluations. The evaluations and the feedback from our domain experts have shown that our solution can significantly improve the maintainability of chatbot knowledge bases.}, author = {Alrik Hausdorf and Lydia M{\"u}ller and Gerik Scheuermann and Andreas Niekler and Daniel Wiegreffe} } @conference {914, title = {Supporting Land Reuse of Former Open Pit Mining Sites using Text Classification and Active Learning}, booktitle = {Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)}, year = {2021}, abstract = {Open pit mines left many regions worldwide inhospitable or uninhabitable. Many sites are left behind in a hazardous or contaminated state, show remnants of waste, or have other restrictions imposed upon them, eg, for the protection of human or nature. Such information has to be permanently managed in order to reuse those areas in the future. In this work we present and evaluate an automated workflow for supporting the post-mining management of former lignite open pit mines in the eastern part of Germany, where prior to any planned land reuse, aforementioned information has to be acquired to ensure the safety and validity of such an endeavor. Usually, this information is found in expert reports, either in the form of paper documents, or in the best case as digitized unstructured text{\textemdash}all of them in German language. However, due to the size and complexity of these documents, any inquiry is tedious and time-consuming, thereby slowing down or even obstructing the reuse of related areas. Since no training data is available, we employ active learning in order to perform multi-label sentence classification for two categories of restrictions and seven categories of topics. The final system integrates optical character recognition (OCR), active-learning-based text classification, and geographic information system visualization in order to effectively extract, query, and visualize this information for any area of interest. Active learning and text classification results are twofold: Whereas the restriction categories were reasonably accurate (> 0.85 F1), the seven topic-oriented categories seemed to be complex even for human annotators and achieved {\textellipsis}}, author = {Christopher Schr{\"o}der and Kim B{\"u}rgl and Yves Annanias and Andreas Niekler and Lydia M{\"u}lller and Daniel Wiegreffe and Christian Bender and Christoph Mengs and Gerik Scheuermann and Gerhard Heyer} } @conference {845, title = {Einsatz von K{\"u}nstlicher Intelligenz (KI) f{\"u}r die Optimierung von Planungsprozessen im Wasserbau}, booktitle = {43. Dresdner Wasserbaukolloquium 2020}, year = {2020}, author = {Patrycja-Jadwiga Sankowska and Nina Kumbruck and Christian Leyh and Andreas Niekler and Daniel Wiegreffe} } @proceedings {840, title = {LocalCompanies: Visual Analytics of spatial aligned regional companies}, year = {2020}, address = {Leipzig}, doi = {10.31219/osf.io/tsdfh}, url = {osf.io/tsdfh}, author = {Alrik Hausdorf and Andreas Niekler and Daniel Wiegreffe} }