scholarly journals SemDaServ: A Systematic Approach for Semantic Data Specification of AI-Based Smart Service Systems

2021 ◽  
Vol 11 (11) ◽  
pp. 5148
Author(s):  
Maurice Preidel ◽  
Rainer Stark
Keyword(s):  
Domain Knowledge ◽  
Systematic Approach ◽  
Service Systems ◽  
Training Data ◽  
Research Approach ◽  
Model Quality ◽  
Smart Services ◽  
Smart Service ◽  
Applicability Evaluation ◽  
Data Specification

To develop smart services to successfully operate as a component of smart service systems (SSS), they need qualitatively and quantitatively sufficient data. This is especially true when using statistical methods from the field of artificial intelligence (AI): training data quality directly determines the quality of resulting AI models. However, AI model quality is only known when AI training can take place. Additionally, the creation of not yet available data sources (e.g., sensors) takes time. Therefore, systematic specification is needed alongside smart service systemsSSS development. Today, there is a lack of systematic support for specifying data relevant to smart services. This gap can be closed by realizing the systematic approach SemDaServ presented in this article. The research approach is based on Blessing’s Design Research Methodology (literature study, derivation of key factors, success criteria, solution functions, solution development, applicability evaluation). SemDaServ provides a three-step process and five accompanying artifacts. Using domain knowledge for data specification is critical and creates additional challenges. Therefore, the SemDaServ approach systematically captures and semantically formalizes domain knowledge in SysML-based models for information and data. The applicability evaluation in expert interviews and expert workshops has confirmed the suitability of SemDaServ for data specification in the context of SSS development. SemDaServ thus offers a systematic approach to specify the data requirements of smart services early on to aid development to continuous integration and continuous delivery scenarios.

scholarly journals Leveraging Circular Economy through a Methodology for Smart Service Systems Engineering

2019 ◽  
Vol 11 (13) ◽  
pp. 3517 ◽  
Author(s):  
Friedrich A. Halstenberg ◽  
Kai Lindow ◽  
Rainer Stark
Keyword(s):  
Systems Engineering ◽  
Circular Economy ◽  
Service Design ◽  
Service Systems ◽  
Service Development ◽  
System Modelling ◽  
Present System ◽  
Service Architecture ◽  
Smart Services ◽  
Smart Service

Product Service Systems (PSS) and Smart Services are powerful means for deploying Circular Economy (CE) goals in industrial practices, through dematerialization, extension of product lifetime and efficiency increase by digitization. Within this article, approaches from PSS design, Smart Service design and Model-based Systems Engineering (MBSE) are combined to form a Methodology for Smart Service Architecture Definition (MESSIAH). First, analyses of present system modelling procedures and systems modelling notations in terms of their suitability for Smart Service development are presented. The results indicate that current notations and tools do not entirely fit the requirements of Smart Service development, but that they can be adapted in order to do so. The developed methodology includes a modelling language system, the MESSIAH Blueprinting framework, a systematic procedure and MESSIAH CE, which is specifically designed for addressing CE strategies and practices. The methodology was validated on the example of a Smart Sustainable Street Light System for Cycling Security (SHEILA). MESSIAH proved useful to help Smart Service design teams develop service-driven and robust Smart Services. By applying MESSIAH CE, a sustainable Smart Service, which addresses CE goals, has been developed.

The different shades of innovation emergence in smart service systems: the case of Italian cluster for aerospace technology

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Orlando Troisi ◽  
Anna Visvizi ◽  
Mara Grimaldi
Keyword(s):  
New Technologies ◽  
Service Systems ◽  
Constructivist Grounded Theory ◽  
Structured Interviews ◽  
Social Rules ◽  
Content Type ◽  
Novel Technologies ◽  
Aerospace Technology ◽  
Smart Service ◽  
Relational Strategies

Purpose The purpose of this paper is to explore the emergence of innovation in smart service systems to conceptualize how actor’s relationships through technology-enabled interactions can give birth to novel technologies, processes, strategies and value. The objectives of the study are: to detect the different enablers that activate innovation in smart service systems; and to explore how these can lead dynamically to the emergence of different innovation patterns. Design/methodology/approach The empirical research adopts an approach based on constructivist grounded theory, performed through observation and semi-structured interviews to investigate the development of innovation in the Italian CTNA (Italian acronym of National Cluster for Aerospace Technology). Findings The identification and re-elaboration of the novelties that emerged from the analysis of the Cluster allow the elaboration of a diagram that classifies five different shades of innovation, introduced through some related theoretical propositions: technological; process; business model and data-driven; social and eco-sustainable; and practice-based. Originality/value The paper embraces a synthesis view that detects the enabling structural and systems dimensions for innovation (the “what”) and the way in which these can be combined to create new technologies, resources, values and social rules (the “how” dimension). The classification of five different kinds of innovation can contribute to enrich extant research on value co-creation and innovation and can shed light on how given technologies and relational strategies can produce varied innovation outcomes according to the diverse stakeholders engaged.

Semi-Supervised Aspect-Based Sentiment Analysis for Case-Related Microblog Reviews Using Case Knowledge Graph Embedding

2021 ◽  
pp. 2050012
Author(s):  
Peilian Zhao ◽  
Cunli Mao ◽  
Zhengtao Yu
Keyword(s):  
Sentiment Analysis ◽  
Domain Knowledge ◽  
Opinion Mining ◽  
Data Augmentation ◽  
Training Data ◽  
Knowledge Graph ◽  
Fine Grained ◽  
Learning Framework ◽  
Proposed Model ◽  
Real World Applications

Aspect-Based Sentiment Analysis (ABSA), a fine-grained task of opinion mining, which aims to extract sentiment of specific target from text, is an important task in many real-world applications, especially in the legal field. Therefore, in this paper, we study the problem of limitation of labeled training data required and ignorance of in-domain knowledge representation for End-to-End Aspect-Based Sentiment Analysis (E2E-ABSA) in legal field. We proposed a new method under deep learning framework, named Semi-ETEKGs, which applied E2E framework using knowledge graph (KG) embedding in legal field after data augmentation (DA). Specifically, we pre-trained the BERT embedding and in-domain KG embedding for unlabeled data and labeled data with case elements after DA, and then we put two embeddings into the E2E framework to classify the polarity of target-entity. Finally, we built a case-related dataset based on a popular benchmark for ABSA to prove the efficiency of Semi-ETEKGs, and experiments on case-related dataset from microblog comments show that our proposed model outperforms the other compared methods significantly.

scholarly journals Mineralization of the surface waters in the Visocica drainage basin: Supplement for the landscape-ecologic analysis

2009 ◽  
Vol 89 (4) ◽  
pp. 141-160 ◽  
Author(s):  
Sanja Mustafic ◽  
Predrag Manojlovic ◽  
Miroljub Milincic
Keyword(s):  
Surface Waters ◽  
Systematic Approach ◽  
Drainage Basin ◽  
The State ◽  
Research Approach ◽  
Geologic Structure ◽  
Spatial Unit ◽  
Water Accumulation ◽  
State Of The Environment

The drainage basin is spatially and functionally clearly defined and relevant hydrologic, geomorphologic and ecologic landscape totality. Therefore, it mostly represents basic geo-spatial unit of generation, monitoring, and studying numerous physical-geographical and geo-ecologic occurrences and processes. One of the most important components of geo-space, on the level of basin, is manifested through the state and quality of surface waters. So, the acceptance of systematic approach in studying mineralization of the surface waters would contribute to the deeper understanding of the process in complex systematic surroundings which drainage basin represents. The Visocica Drainage Basin was chosen as proving ground of this kind of the research approach for several reasons. The highest specific runoff on the territory of Eastern Serbia, heterogeneous geologic structure of terrain, almost complete absence of the influence of the anthropogenic factor on the state of the environment, as well as the existence of water accumulation enabled perception of the values of dissolved mineral substances of surface waters as landscape-ecologic component of geo-space.

scholarly journals Monte Carlo and Reconstruction Membership Inference Attacks against Generative Models

2019 ◽  
Vol 2019 (4) ◽  
pp. 232-249 ◽  
Author(s):  
Benjamin Hilprecht ◽  
Martin Härterich ◽  
Daniel Bernau
Keyword(s):  
Data Privacy ◽  
Information Leakage ◽  
Generative Models ◽  
Generative Model ◽  
Training Data ◽  
Generative Adversarial Networks ◽  
Data Sets ◽  
Success Rates ◽  
Model Quality ◽  
Type Formalization

Abstract We present two information leakage attacks that outperform previous work on membership inference against generative models. The first attack allows membership inference without assumptions on the type of the generative model. Contrary to previous evaluation metrics for generative models, like Kernel Density Estimation, it only considers samples of the model which are close to training data records. The second attack specifically targets Variational Autoencoders, achieving high membership inference accuracy. Furthermore, previous work mostly considers membership inference adversaries who perform single record membership inference. We argue for considering regulatory actors who perform set membership inference to identify the use of specific datasets for training. The attacks are evaluated on two generative model architectures, Generative Adversarial Networks (GANs) and Variational Autoen-coders (VAEs), trained on standard image datasets. Our results show that the two attacks yield success rates superior to previous work on most data sets while at the same time having only very mild assumptions. We envision the two attacks in combination with the membership inference attack type formalization as especially useful. For example, to enforce data privacy standards and automatically assessing model quality in machine learning as a service setups. In practice, our work motivates the use of GANs since they prove less vulnerable against information leakage attacks while producing detailed samples.

Sign in / Sign up

Export Citation Format

Share Document