A5075284697- Natural Language Processing Techniques
- Speech and dialogue systems
- Linguistic research and analysis
- Semantic Web and Ontologies
- Linguistic Education and Pedagogy
- Multi-Agent Systems and Negotiation
- Topic Modeling
- AI in Service Interactions
- Business Process Modeling and Analysis
- Hand Gesture Recognition Systems
- Linguistics, Language Diversity, and Identity
- Social Robot Interaction and HRI
- Syntax, Semantics, Linguistic Variation
- Model-Driven Software Engineering Techniques
- Text and Document Classification Technologies
- Service-Oriented Architecture and Web Services
- Constraint Satisfaction and Optimization
- Flexible and Reconfigurable Manufacturing Systems
- Multimodal Machine Learning Applications
- Lexicography and Language Studies
- Human Pose and Action Recognition
- Spam and Phishing Detection
- Video Analysis and Summarization
- Text Readability and Simplification
- Advanced Database Systems and Queries
German Research Centre for Artificial Intelligence
2005-2022
Deutsches Forschungsnetz
1991-2015
Klinikum Saarbrücken
2007
Saarland University
1990-1996
Technische Universität Berlin
1986-1988
Universität Hamburg
1983
Customer care in technical domains is increasingly based on e-mail communication, allowing for the reproduction of approved solutions. Identifying customer's problem often time-consuming, as space changes if new products are launched. This paper describes a approach to classification requests shallow text processing and machine learning techniques. It implemented within an assistance system call center agents that used commercial setting.
Current work in surface realization concentrates on the use of general, abstract algorithms that interpret large, reversible grammars. Only little attention has been paid so far to many small and simple applications require coverage a sublanguage at different degrees sophistication. The system TG/2 described this paper can be smoothly integrated with deep generation processes, it integrates canned text, templates, context-free rules into single formalism, allows for both textual tabular...
Appointment scheduling is a problem faced daily by many individuals and organizations. Cooperating agent systems have been developed to partially automate this task. In order extend the circle of participants as far possible we advocate use natural language transmitted email. We describe COSMA, fully implemented German server for existing appointment systems. COSMA can cope with multiple dialogues in parallel, accounts differences dialogue behaviour between human machine agents. NL coverage...
The natural language system DISCO is described. It combines• a powerful and flexible grammar development system;• linguistic competence for German including morphology, syntax semantics;• new methods performance modeling on the basis of high-level grammars;• modelling multi-agent dialogue competence;• an interesting sample application appointment scheduling calendar management.
This paper is based on several attempts to provide a definition of an Artificial Companion that can be found in the referenced literature.Although accepted by research community, such descriptions set very high expectations agents, but they do not address technical feasibility and system limitations, why need elaborated precise explanation.In this paper, we focus computer agents simulate human language behaviour, are aimed serve, assist accompany their owner over long period time, call...
In order to support the efficient development of NL generation systems, two orthogonal methods are currently pursued with emphasis: (1) reusable, general, and linguistically motivated surface realization components, (2) simple, task-oriented template-based techniques. this paper we argue that, from an application-oriented perspective, benefits both still limited. improve situation, suggest evaluate shallow associated increased flexibility. We advise a close connection between...
In this paper we describe a design approach for an Artificial Conversational Companion according to earlier identified requirements of utility, adaptivity, conversational capabilities and long-term interaction. The is aimed help advanced learners foreign language practice conversation via instant messenger dialogues. order model meaningful interaction with application case, it necessary understand how natural chat between human experts works. For purpose, created corpus from messenger-based...
This paper describes part of a novel view planning the assembly cars at shop floor, which is currently being explored in EU-funded project INTERACT, 3D worker simulations are automatically generated from textual descriptions. Under this view, all carried out virtually, thereby interactively exploiting workers' knowledge. We suggest solutions to subtask mapping descriptions onto motion sequences. Consider description like "Tighten arm support with cordless screw driver on center console"....
Using the formalism of generalized phrase structure grammar (GPSG) in an NL system (e.g. for machine translation (MT) is promising since modular very well suited to meet some particular needs MT. However, it seems impossible implement GPSG its 1985 version straightforwardly. This would involve a vast overgeneration structures as processes filter out everything but admissible tree(s). We therefore argue constructive where information gathered subsequent steps produce syntactic structures. As...
In everyday life, Deaf People face barriers because information is often only available in spoken or written language. Producing sign language videos showing a human interpreter not feasible due to the amount of data required changes frequently. The ongoing AVASAG project addresses this issue by developing 3D avatar for automatic translation texts into public services. trained using recordings interpreters translating text For purpose, we create corpus with video and motion capture an...
This paper presents a novel information system integrating advanced extraction technology and automatic hyper-linking. Extracted entities are mapped into domain ontology that relates concepts to selection of hyperlinks. For extraction, we use SProUT, generic platform for the development multilingual text processing components. By combining finite-state unification-based formalisms, grammar formalism used in SProUT offers both efficiency high degree decalrativeness. The ExtraLink demo...
We discuss the use of NL systems in domain appointment scheduling. Appointment scheduling is a problem faced daily by many people and organizations, typically solved using communication natural language. In general, cooperative interaction between several participants required whose calendar data are distributed rather than centralized. this multi-agent environment, makes it possible for machines humans to cooperate solving problems. describe COSMA (Cooperative Schedule Managament Agent)...
Abstract This contribution describes the German EU Council Presidency Translator (EUC PT), a machine translation service created for in second half of 2020, which is open to general public. Following series earlier presidency translators, version exhibits important extensions and improvements. The EUC PT first integrate systems from commercial vendors, public services, research center, using mix custom generic engines, introduce new webpage widget. A further feature close collaboration with...