Executable Knowledge Graph for Transparent Machine Learning in Welding Monitoring at Bosch

Abstract

With the development of Industry 4.0 technology, modern industries such as Bosch's welding monitoring witnessed the rapid widespread of machine learning (ML) based data analytical applications, which in the case of welding monitoring has led to more efficient and accurate welding monitoring quality. However, industrial ML is affected by the low transparency of ML towards non-ML experts needs. The lack of understanding by domain experts of ML methods hampers the application of ML methods in industry and the reuse of developed ML pipelines, as ML methods are often developed in an ad hoc manner for specific problems. To address these challenges, we propose the concept and a system of executable Knowledge Graph (KG), which formally encode ML knowledge and solutions in KGs, which serve as common language between ML experts and non-ML experts, thus facilitate their communication and increase the transparency of ML methods. We evaluated our system extensively with an industrial use case at Bosch, showing promising results.Speaker's Bio: Zhuoxun Zheng is a Data Scientist and Doctoral Researcher at Bosch Center for Artificial Intelligence. He obtained his Master's degree in Mechanical Engineering at Karlsruhe Institute of Technology. Currently, he is working on knowledge enhanced machine learning and neuro-symbolic reasoning to increase transparency, trust, and performance of ML.Relevancy to CIKM: Knowledge enhanced ML in the industry.

ExeKG: Executable Knowledge Graph System for User-friendly Data Analytics

“…Based on the GUI, users are able to construct executable KGs in three ways [14]: creation, modification and integration. Creation refers to represent specific data analytics pipelines by instantiating templates from the scratch, choosing the appropriate template which determines the domain and structure of the KG, and filling variables of entities and properties guided by the GUI step by step (Fig.…”

ScheRe: Schema Reshaping for Enhancing Knowledge Graph Construction

“…And common Queries are selected by the users. The KGs in the KG Data Layer then can be used for applications like Query-Based Analytics [22] and ML Analytics [19,21,23,26] in the Application Layer.…”

Addressing the Scalability Bottleneck of Semantic Technologies at Bosch

“…When the welding is performed during car body manufacturing, each such body has up to 6.000 welding spots, generating a large amount of data instances. The data is distributed across several analytical pipelines in charge of feed statistics, training traditional ML models, quality control measures, and many more [8].…”