Accurate estimation of polyp size during colonoscopy is critical for risk assessment and surveillance planning in colorectal cancer prevention. However, current methods often rely on subjective visual judgment, leading to inconsistencies and potential misclassification. This study proposes a deep learning framework that enables automated and objective polyp size classification by integrating RGB and depth information. The approach leverages a modified Af-SfM module to generate refined and rectified depth maps, which are combined with RGB inputs to support classification into clinically relevant size categories. The model was trained and validated on a dataset of over 10,000 annotated colonoscopic images curated by expert gastroenterologists. Experimental results demonstrate that incorporating rectified depth information significantly improves classification performance over RGB-only baselines. For polyps measuring 10 mm or larger, the system achieved a precision of 91.5% and a recall of 93.6%. These findings highlight the potential of depth-enhanced deep learning methods to support more consistent and accurate polyp size estimation in clinical endoscopy.

Introducing ADEA: a German dataset that captures online dialogues and focuses on ethical issues related to future AI applications. This dataset, which includes over 2800 labeled user utterances on four different topics, is specifically designed for the training of chatbots that can navigate the complexities of real-world ethical AI conversations. The creation of these dialogues is the result of two carefully conducted studies in which university students interacted with an argumentative dialogue system. A fundamental part of our methodology is the use of German argument graphs. These graphs not only form the knowledge base of the dialogue system but also serve as an effective annotation scheme for the dialogues. Apart from the introduction of the dataset and the argument graphs, we provide a preliminary benchmark using GPT-4 via the OpenAI API. This provides researchers with a concrete reference point while demonstrating the potential of our dataset. We make our dataset and argument graphs available at https://github.com/HaupChris/ADEA-Dialogue-Dataset.

Abstract This paper explores the potential of a German-language chatbot to engage users in argumentative dialogues on ethically sensitive topics. Utilizing an argumentative knowledge graph, the chatbot is equipped to engage in discussions on the ethical implications of autonomous AI systems in hypothetical future scenarios in the fields of medicine, law, and self-driving cars. In a study with 178 student participants, we investigated the chatbot’s argumentation effect—its ability to offer new perspectives, gain user acceptance, and broaden users’ viewpoints on complex issues. The results indicated a substantial argumentation effect, with 13–21% of participants shifting their opinions to more moderate stances after interacting with the chatbot. This shift demonstrates the system’s effectiveness in fostering informed discourse and increasing users’ understanding of AI ethics. While the chatbot was well-received, with users acknowledging the quality of its arguments, we identified opportunities for improvement in its argument recognition capabilities. Despite this, our results indicate the chatbot’s potential as an educational tool in engaging users with the ethical dimensions of AI technology and promoting informed discourse.

Abstract An important but very time consuming part of the research process is literature review. An already large and nevertheless growing ground set of publications as well as a steadily increasing publication rate continue to worsen the situation. Consequently, automating this task as far as possible is desirable. Experimental results of systems are key-insights of high importance during literature review and usually represented in form of tables. Our pipeline KIETA exploits these tables to contribute to the endeavor of automation by extracting them and their contained knowledge from scientific publications. The pipeline is split into multiple steps to guarantee modularity as well as analyzability, and agnosticim regarding the specific scientific domain up until the knowledge extraction step, which is based upon an ontology. Additionally, a dataset of corresponding articles has been manually annotated with information regarding table and knowledge extraction. Experiments show promising results that signal the possibility of an automated system, while also indicating limits of extracting knowledge from tables without any context.

Albin Zehe, Leonard Konle, Lea Katharina Dümpelmann, Evelyn Gius, Andreas Hotho, Fotis Jannidis, Lucas Kaufmann, Markus Krug, Frank Puppe, Nils Reiter, Annekea Schreiber, Nathalie Wiedmer. Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume. 2021.

This paper examines the automatic transcription of lyrics of Medieval music manuscripts by applying recent developments in the area of Automatic Text Recognition (ATR). We evaluate the performance of a CNN/LSTM-network on five different manuscripts dating from the 12th to 16thcentury and examine the impact of two different line segmentation approaches: Using an accurate manual segmentation yielded a Character Error Rate (CER) of up to 6.7% whereas 8.2% were reached on a fully automatic one. Furthermore, we propose an algorithm for the assignments of syllables to their respective neume by finding valid matches based on the positional output of the ATR. Depending on the ATR accuracy, an F1-score of over 99% was obtained.

In this paper we introduce a method that significantly reduces the character error rates for OCR text obtained from OCRopus models trained on early printed books. The method uses a combination of cross fold training and confidence based voting. After allocating the available ground truth in different subsets several training processes are performed, each resulting in a specific OCR model. The OCR text generated by these models then gets voted to determine the final output by taking the recognized characters, their alternatives, and the confidence values assigned to each character into consideration. Experiments on seven early printed books show that the proposed method outperforms the standard approach considerably by reducing the amount of errors by up to 50% and more.

We propose a high-performance fully convolutional neural network (FCN) for historical document segmentation that is designed to process a single page in one step. The advantage of this model beside its speed is its ability to directly learn from raw pixels instead of using preprocessing steps e. g. feature computation or superpixel generation. We show that this network yields better results than existing methods on different public data sets. For evaluation of this model we introduce a novel metric that is independent of ambiguous ground truth called Foreground Pixel Accuracy (FgPA). This pixel based measure only counts foreground pixels in the binarized page, any background pixel is omitted. The major advantage of this metric is, that it enables researchers to compare different segmentation methods on their ability to successfully segment text or pictures and not on their ability to learn and possibly overfit the peculiarities of an ambiguous hand-made ground truth segmentation.

Abstract Rule-based information extraction is an important approach for processing the increasingly available amount of unstructured data. The manual creation of rule-based applications is a time-consuming and tedious task, which requires qualified knowledge engineers. The costs of this process can be reduced by providing a suitable rule language and extensive tooling support. This paper presents UIMA Ruta, a tool for rule-based information extraction and text processing applications. The system was designed with focus on rapid development. The rule language and its matching paradigm facilitate the quick specification of comprehensible extraction knowledge. They support a compact representation while still providing a high level of expressiveness. These advantages are supplemented by the development environment UIMA Ruta Workbench. It provides, in addition to extensive editing support, essential assistance for explanation of rule execution, introspection, automatic validation, and rule induction. UIMA Ruta is a useful tool for academia and industry due to its open source license. We compare UIMA Ruta to related rule-based systems especially concerning the compactness of the rule representation, the expressiveness, and the provided tooling support. The competitiveness of the runtime performance is shown in relation to a popular and freely-available system. A selection of case studies implemented with UIMA Ruta illustrates the usefulness of the system in real-world scenarios.

Regarding knowledge-based systems (KBS), the once common paradigm—perfectly mimicking the abilities of human experts—is gradually replaced by an increasing demand for more (active) user participation. The therefore required key features intelligent data input handling and results presentation/justification, however, are still treated separately most often. In this paper, we propose a novel KBS paradigm, the Clarification KBS, as a mash up type of consultation and justification interaction and presentation—intended to foster active user participation according to users’ competency; the KBS’ explicability; and support for learnability. We discuss both the theoretical concept of clarification KBS, as well as several distinct variants for realizing appropriate UI- and interaction designs. Further, we report diverse evaluation experiences regarding a specific instantiation of clarification KBS—the ITree KBS type; therefore, we subsume its specific characteristics, describe corresponding case studies, and discuss the respective results.

Subgroup discovery (also known as Pattern Mining or Supervised Descriptive Rule Discovery) searches for descriptions of subsets in a dataset that differ from the total population with respect to a given target concept. In this paper we argue that in the traditional approach potentially interesting complex patterns with an unexpected relative increase of the target share remain undiscovered while on the other hand less surprising patterns are returned. Therefore, we present a generalized approach on subgroup discovery, in which the target share in the subgroup is not compared to the target share in the total population, but to the expectations a user has given the knowledge of more general (simpler) patterns. We claim that the resulting complex patterns are more interesting for the user and are less biased towards simpler patterns with a positive influence on the target concept. In order to estimate these expectations we utilize local models, i.e., fragments of Bayesian Networks. The proposed approach is evaluated using data from the UCI repository as well as on two totally different real world applications that investigate university student drop-out rates and identify spammers in a social book marking system.

Text-based authoring using knowledge markups is an increasingly popular editing paradigm in manual knowledge acquisition. Closed world authoring environments support the user to form a coherent knowledge base by checking the referenced objects against a set of declared domain objects. In this scenario, the task of efficient translation (compilation) of the text sources is non-trivial. Additionally, in real-world applications frequent small changes are performed on the source documents and instant feedback to the author is crucial. Therefore, a scalable compilation into the target knowledge representations is necessary. In this paper, we introduce a general algorithm for the incremental compilation of knowledge documents, that analyzes the current document modifications and performs minimal updates on the knowledge base. We provide a formal proof of the correctness of the algorithm and show the effectiveness of the approach in several case studies, using various kinds of knowledge representations and markups.

This paper presents the knowledge-system SONOCONSULT – an intelligent system in the medical domain. We evaluated the accuracy, acceptance and impact of SONOCONSULT, which has been used in clinical routine since 2002. The system was well accepted and had a significant clinical impact. In contrast to our original expectations, the diagnostic conclusions, although inferred with high accuracy, were less important especially for experienced physicians.

This paper presents a semi-automatic approach for confounding-aware subgroup discovery: We present a method that provides the means for detecting potentially confounded subgroup patterns, other unconfounded relations, and/or patterns that are affected by effect- modification. Since there is no purely automatic test for confounding, the discovered relations are subsequently presented to the user in a semi-automatic approach. Furthermore, we show how to utilize (causal) domain knowledge for improving the results of the algorithm, since confounding is itself a causal concept. The applicability and benefit of the presented technique is illustrated by examples from a case-study in the medical domain.

This paper presents an approach for rapid knowledge capture using subgroup-discovery techniques. The method enables the acquisition of scoring rules - a knowledge representation that is easy to understand and to maintain. Furthermore, the method features an incremental refinement step that can be applied for fine-tuning of the learned relations. We provide a case study demonstrating the applicability of the presented method using a knowledge base from the biological domain.

This paper presents a collaborative knowledge engineering approach based on the widespread wiki technique.The interface of a standard wiki system is extended to allow for the capture, the maintenance and the use of knowledge systems.

User-centered consultation systems can be viewed as systems with trust in the competence of the user, whereas conventional knowledge systems are often based on some mistrust. By offering a variety of formal and informal diagnostic services mediated by flexible agents, the user solves his or her problem by asking the agents for specific information complementary to his or her knowledge. The development process is flexible enough to build minimalist Wiki-like informal systems very quickly, which can be incrementally refined and formalized in a distributed manner. The approach is demonstrated with two case studies.

Although a lot of work in the field of knowledge acquisition has been done, the manual development of diagnostic knowledge systems by domain experts still is a very complex task. In this paper we will present an incremental approach for building diagnostic systems based on set-covering models. We start with a simple model describing the coarse structure between diagnoses and findings. Subsequently, this simple model can be euhanced by similarities, weights and probabilities to increase the accuracy of the knowledge and the resulting system. We will also show how these static set-covering models can be combined with dynamic set-covering models including higher level knowledge about causation effects. We will motivate how dynamic set-covering models can be used for implementing diagnostic systems including therapy effects. Finally, we report on two practical applications dealing with set-covering models from the geo-ecological and from the medical domain, respectively, that we have implemented.

A problem-solving method for assignment problems and a corresponding expert system shell named COKE are presented and evaluated. The method consists of four main steps. Until all elements are assigned: (1) select an element to be assigned next, (2) propose a partner element for that element, (3) if new constraints are violated after the proposed step, try exchanging elements to remove or minimize the constraint violations from a local point of view, and (4) if the completed assignment still violates constraints, try exchanges to remove or minimize them from a global point of view and with more effort. Each step can take advantage of problem-specific knowledge. The evaluation with COKE shows that flexibility is of key importance in dealing with different kinds of assignment problems.>