Driven by the challenges of Industry 4.0 seamless data integration and conversion becomes an even more pressing topic than it already has been in the past. In this context virtual technologies such as Virtual Reality (VR) and Augmented Reality (AR) will play a defining role when it comes down to speed up the development process of new products, factory planning and supporting the workforce with novel intelligent assistance systems to keep up with the faster growing product diversity, especially in the face of the demographic challenge. The key enabling technology to foster this development is the seamless and automated integration of all product data defined in the CAD systems into VR/AR systems. Currently, there is still a major gap in this data integration when more than just plain geometry should be transferred, especially outside of the software ecosphere of a CAD system provider. Considerable effort by VR experts has to be put in to reduce the use case specific complexity of models and to rebuild the already in the CAD system defined kinematic mechanisms and animation. Whilst these obstacles were hindering SMEs in the past to participate of the benefits that virtual technologies provide, this cannot be afforded any longer. In this paper we will outline a CAD and VR system independent workflow for an automated model complexity reduction, animation and kinematic mechanism adoption. We do this by defining light weighted system interfaces and showing solution concepts for each area of the conversion problem. Finally, we verify the presented method on a proof of concept implementation

Named entity recognition (NER) constitutes an important step in the processing of unstructured text content for the extraction of information as well as for the computer-supported analysis of large amounts of digital data via machine learning methods. However, NER often relies on domain-specific knowledge, being conducted manually in a time- and human-resource-intensive process. These can be reduced with statistical models performing NER automatically. The current work investigates whether Conditional Random Fields (CRF) can be efficiently trained for NER in German texts, by means of an iterative procedure combining self-learning with a manual annotation–active learning–component. The training dataset increases continuously with the iterative procedure. Whilst self-learning did not markedly improve the performance of the CRF for NER, the manual annotation of sentences with the lowest probability of correct prediction clearly improved the model F1-score and simultaneously reduced the amount of manual annotation required to train the model. A model with an F1-score of 0.885 was able to be trained in 11.4 h.

Mobile communication devices are a popular means of planning, commissioning and carrying out criminal offenses. In particular, data from messengers such as WhatsApp or Telegram often contain conclusive information. Organized crime also usually involves many devices, but not all of them contain the full history of communication. Rather, it is heavily fragmented due to individual deletions of messages or different joining times to groups. A singular evaluation of individual devices is therefore often not expedient, since important relationships cannot be recognized. Furthermore, communication is often distributed across different channels and modalities and can only be fully and correctly understood through a joint semantic analysis. The linking of related communications of different devices enables an almost complete reconstruction of the communication with a simultaneous reduction in reading effort by merging identical messages. Grouping coherent messages into conversations enables efficient comparison with a knowledge model. Building such a model is complex, but can be supported by a term recommender system. In this paper, MoNA is presented as a platform that implements these approaches and enables an assisted analysis of mobile communications.

The decline in coral reef health presents a complex management issue. While several causes of decline have been identified and are under continued study, it is often difficult to discern management actions necessary to address multiple near- and far-field stressors to these ecosystems. As a result, resource managers seek tools to improve the understanding of ecosystem condition and to develop management responses to reduce local and regional pressures in the wake of larger, global impacts. A research study conducted from 2010 to 2014 in southeast Florida, USA consisted of two objectives: (1) conduct a needs assessment survey with coral reef and marine resource managers to identify data needs and the preferred design and delivery of climate information; and (2) develop and evaluate prototype decision support tools. The needs assessment process was helpful for identifying the types of climate information managers would like to obtain to inform decision making and to specify the preferred format for the delivery of that information. Three prototype tools were evaluated by managers using pre/post surveys that included hands-on tutorials to explore the functionality of each. Manager responses were recorded using a five-point scale with 1 being No or Not Useful to 5 being Absolutely or Very Useful. The median responses rated the usefulness of the tools (4), if they would consider using the tool (4), and if they would recommend using the tool to other managers (4 or 5). The median response for increasing manager’s knowledge about climate impacts after completing a tutorial of each of the climate tools was a 3 (moderately useful). Of the managers surveyed in the pre/post-survey, all but one stated they believed they would use the decision support tools in the future with the single response due to wealth of data availability in their institution

The digital transformation of higher education demands effective and efficient methods for learning assessment. Research has shown that tests have the potential to be an appropriate medium not only to diagnose learning success but to improve learning outcomes/processes in general. In this study, we examine possible connections between the use of digital tests and learning success. Based on a survey (n=71) and a quantitative analysis (n=214) with logging and exam assessment data, we find three points that indicate positive correlation: Firstly, students perceive digital tests as motivating and useful. Secondly, we detect a positive relation between the points students achieved in the online tests and the exam results (Spearman correlation coefficient of 0.39), and thirdly, students who participated continually in the digital testing system on average succeeded and passed the exam. The results show the importance of systematic digital testing in learning management systems, but also the need for more research with respect to other contexts and domains.