This study investigates the dynamics of user interactions within the gaming community surrounding Genshin Impact, particularly during a backlash over accusations of racism and cultural appropriation following a July 2024 update. By analysing comments and interaction patterns on Reddit, this study employs a feature extraction method based on the TF-IDF concept, enhanced by voting counts, to better reflect users' interests and the nuances of community discourse. The findings reveal significant variances in user characteristics and subgroup interactions, providing insights into the demographic and linguistic variations in culture-based communities on social media.

Developing metadata schema on research data requires not only the specialized domain knowledge but also in-depth knowledge about semantic web technologies, which makes the task difficult for those who are not familiar with them. To solve the problem, we aim to develop a metadata design system for researchers and collaborators with various knowledge, skills and expertise to co-design metadata schema by cross-organizational collaboration. In this paper, we describe the basic design concept, overview, and future issues of the initial system we designed.

Knowledge Graph Embedding (KGE) is a method of mapping knowledge graphs, which represent real-world entities and the connections between such entities, into a low-dimensional vector space and is used for similar node search and link prediction. Uncertain Knowledge Graph Embedding (UKGE) can obtain embeddings with a confidence score, including ambiguous and uncertain expressions. However, these KGE models are often black boxes; hence, interpreting the knowledge that contributed to link prediction is difficult. In this paper, we propose a method to extend Kelpie, an explanation framework for knowledge graph embedding, for UKGE.

In this paper, we extend a tool for extracting knowledge graph candidates by leveraging dependency structure analysis. Specifically, instead of using a dependency structure analysis tool, we replace the syntactic relationship determination based on the derivation rules of one of the wellknown parsing algorithms, the CKY algorithm, with a prompt-based determination using a Large Language Model (LLM). This approach allows a better integration of knowledge graph extraction with syntactic structure analysis. The prompts focus on bunsetsus, the basic units of meaning in Japanese, allowing us to effectively address domain-specific writing styles and handle named entities by restricting sentence sets.

The large language model (LLM) based recommendation system is effective for sequential recommendation and knowledge graph-based prompt tuning has been proposed. However, it is also necessary to consider user knowledge; thus, we propose a user knowledge prompt, which converts a user knowledge graph into a prompt. We conducted experiments on two types of datasets (movie and music) and show the improvement of our proposed user knowledge prompt.

To identify the causes of defects in manufacturing systems, past maintenance records are important materials to utilize. Previous studies have attempted to construct a knowledge graph for utilizing these maintenance records. However, challenges have included the inconsistency and specificity of the vocabulary in the maintenance records, and the high cost of machine learning for constructing the knowledge graph. This study proposes a method for constructing a knowledge graph from maintenance records using a pre-trained large language model and a small domain ontology. In constructing a knowledge graph based on actual maintenance records, the proposed method has linked and structured a number of events. In the future, we plan to work on defect cause inference using the knowledge graph.

The maintenance activities of automated manufacturing lines require a deep understanding of equipment and skills. One widely used failure analysis method is Failure Mode and Effect Analysis (FMEA). Previous research has attempted to utilize knowledge of existing failure phenomena based on past analyses to assist in identifying causes of failures. However, these studies have been limited to retrieving descriptions from knowledge bases and have not extended to searching for similar failure phenomena according to differences in manufacturing line processes. This study proposes a method to infer failure causes using embedding techniques by constructing a knowledge graph that structures the conceptual differences in manufacturing lines from FMEA descriptions.