Violence detection explanation via semantic roles embeddings

BMC Medical Informatics and Decision Making

Davide Colla¹  Daniele P. Radicioni¹  Enrico Mensa¹  Marco Giustini²  Alessio Pitidis³  Marco Dalmasso⁴  Carlo Mamo⁴ 

[1] Department of Computer Science, University of Turin, Corso Svizzera 185, 10149, Turin, Italy;Reparto Epidemiologia ambientale e sociale Dipartimento Ambiente e Salute (DAMSA) Istituto Superiore di Sanità, Viale Regina Elena, 299, 00161, Roma, Italy;Reparto Epidemiologia ambientale e sociale Dipartimento Ambiente e Salute (DAMSA) Istituto Superiore di Sanità, Viale Regina Elena, 299, 00161, Roma, Italy;Data Analysis Services, B2C Innovation Inc. - Digital Services, Corso Magenta 69/A, PO Box 20123, Milan, Italy;Servizio sovrazonale di Epidemiologia dell’ASL TO3 della Regione Piemonte, Via Sabaudia 164, 10095, Grugliasco (TO), Italy;

关键词: XAI;    Explanation;    Text categorization;    Categorization explanation;    Word embeddings;    Semantic frames;    Slot filling;    Event extraction;    Violent event tracking;   

DOI  :  10.1186/s12911-020-01237-4

来源: Springer

【 摘 要 】

BackgroundEmergency room reports pose specific challenges to natural language processing techniques. In this setting, violence episodes on women, elderly and children are often under-reported. Categorizing textual descriptions as containing violence-related injuries (V) vs. non-violence-related injuries (NV) is thus a relevant task to the ends of devising alerting mechanisms to track (and prevent) violence episodes.MethodsWe present ViDeS (so dubbed after Violence Detection System), a system to detect episodes of violence from narrative texts in emergency room reports. It employs a deep neural network for categorizing textual ER reports data, and complements such output by making explicit which elements corroborate the interpretation of the record as reporting about violence-related injuries. To these ends we designed a novel hybrid technique for filling semantic frames that employs distributed representations of terms herein, along with syntactic and semantic information. The system has been validated on real data annotated with two sorts of information: about the presence vs. absence of violence-related injuries, and about some semantic roles that can be interpreted as major cues for violent episodes, such as the agent that committed violence, the victim, the body district involved, etc.. The employed dataset contains over 150K records annotated with class (V,NV) information, and 200 records with finer-grained information on the aforementioned semantic roles.ResultsWe used data coming from an Italian branch of the EU-Injury Database (EU-IDB) project, compiled by hospital staff. Categorization figures approach full precision and recall for negative cases and.97 precision and.94 recall on positive cases. As regards as the recognition of semantic roles, we recorded an accuracy varying from.28 to.90 according to the semantic roles involved. Moreover, the system allowed unveiling annotation errors committed by hospital staff.ConclusionsExplaining systems’ results, so to make their output more comprehensible and convincing, is today necessary for AI systems. Our proposal is to combine distributed and symbolic (frame-like) representations as a possible answer to such pressing request for interpretability. Although presently focused on the medical domain, the proposed methodology is general and, in principle, it can be extended to further application areas and categorization tasks.

【 授权许可】

CC BY