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.

Automated cybergrooming detection can support the review of large volumes of online communication, but public datasets and detection approaches remain predominantly English-centered. This study presents a reproducible workflow for constructing target-language cybergrooming datasets and detection models from English resources, demonstrated for German. To the best of our knowledge, this is the first published study to construct and systematically evaluate German-language cybergrooming datasets and detection models. Four publicly available English cybergrooming datasets were converted into a shared structure and translated into informal German chat language. Translation quality was estimated using an aggregate of COMETKiwi and MetricX scores termed CometricX. Low-quality messages were retranslated and reevaluated. Cybergrooming messages were additionally augmented through German–Catalan backtranslation. The resulting datasets, with and without augmentation, were used to train and evaluate embedding-based linear support vector machines (SVMs) and ModernBERT models. Evaluation comprised translated test data and 156 manually labeled German conversations from five closed law-enforcement cases. The Perverted Justice Zig ChitChat dataset (PJZC) achieved the highest aggregate quality estimate and the most robust detection performance on real-world German conversations. In contrast, performance on machine-translated test data overestimated real-world detection performance. Furthermore, backtranslation yielded only limited and inconsistent improvements. Overall, the proposed workflow provides a practical starting point for developing target-language cybergrooming detectors while highlighting careful source dataset selection and evaluation on authentic target-language data.