Case Characterization and Discipline Measurement Method Based on Discipline Inspection and Supervision Knowledge Graph

Abstract

As an important link to realize intelligent discipline inspection and supervision, the term “case characterization and discipline measurement” refers to the automatic extraction of material facts from case description and the conclusion of conformity and nonconformity after comparison in accordance with legal norms. In response to the problem that there was no special method for the task of case characterization and discipline measurement, the paper combined the practical case handling process of the staff and proposed a method of case characterization and discipline measurement based on discipline inspection and supervision knowledge graph. The method uses the knowledge graph as auxiliary information and aligns the entities of regulations and cases using knowledge fusion technology to construct the discipline inspection and supervision knowledge graph. For the newborn case descriptions, named entity recognition technology is used to extract the key elements that determine the verdict outcome. Similar cases were identified with the same discipline breach nature. Then, text classification technology is used to predict the severity of case circumstances. Combined with the disciplinary violation facts, the disciplinary result is given according to the party discipline rules. Experiments were carried out with a dataset of typical cases notified by the discipline inspection and supervision. According to the experimental results, the proposed method shows its validity, which improves the interpretability of case characterization and discipline measurement and fills the field gap.