Foundational Study on Authorship Attribution of Japanese Web Reviews for Actor Analysis

ArXi:2604.16376v1 Announce Type: new This study investigates the applicability of authorship attribution based on stylistic features to actor analysis in threat intelligence. As a foundational step toward future application to dark web forums, we conducted experiments using Japanese review data from clear web sources. We constructed datasets from Rakuten Ichiba reviews and compared four methods: TF-IDF with logistic regression (TF-IDF+LR), BERT embeddings with logistic regression (BERT-Emb+LR), BERT fine-tuning (BERT-FT), and metric learning with $k$-nearest neighbors (Metric+kNN.