Scholarlink

The problem addressed in this study is the limitation of traditional criminal investigation methods in analyzing large volumes of textual data from mobile devices, especially in contexts where the language used contains slang, spelling errors, and deliberate strategies to hinder detection. Therefore, the goal is to develop and demonstrate an approach based on text miningText mining techniques and the use of Semantic Web ontologies to improve the interpretation and organization of forensic data, enhancing the efficiency and accuracy of investigations. The methodology involved extracting WhatsApp’s databases using forensic techniques such as apk downgrade, followed by visualization and manipulation via DB Browser for SQLite. The data were mapped into RDF in GraphDB, using FOAF and Schema.org ontologies, integrating spelling correction procedures, a dictionary of specific terms, and similarity verification. Additionally, SPARQL queries were applied to structure and correlate relevant information, enabling inferences about connections between individuals and events. The results indicate that the combined application of text miningText mining and ontologies enables the identification of behavioral patterns, reconstruction of eventsCriminal event reconstruction, and detection of hidden relationships between interlocutors. The approach was able to recover deleted messages through the merging of backups and to identify critical content, even when it was masked by coded terms or abbreviations. Practical cases demonstrated that the proposed tool offers a low-cost, easy-to-use alternative for non-specialist investigators, while maintaining the technical robustness required for forensic use. The integration of text miningText mining, ontologies, and Semantic Web technologies represents a significant advance for criminal investigations by enabling the structured and contextualized treatment of heterogeneous data. This strategy enhances decision-making, reduces analysis time, and increases the predictive capacity of investigations, contributing to more effective promotion of justice.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Criminal Investigations, Use of Text Mining and Ontologies

  • 章节

  • 发表日期：

    2026年

查看全文
  • Rogerio Atem de Carvalho,
  • Petterson Faria de Souza

摘要

The problem addressed in this study is the limitation of traditional criminal investigation methods in analyzing large volumes of textual data from mobile devices, especially in contexts where the language used contains slang, spelling errors, and deliberate strategies to hinder detection. Therefore, the goal is to develop and demonstrate an approach based on text miningText mining techniques and the use of Semantic Web ontologies to improve the interpretation and organization of forensic data, enhancing the efficiency and accuracy of investigations. The methodology involved extracting WhatsApp’s databases using forensic techniques such as apk downgrade, followed by visualization and manipulation via DB Browser for SQLite. The data were mapped into RDF in GraphDB, using FOAF and Schema.org ontologies, integrating spelling correction procedures, a dictionary of specific terms, and similarity verification. Additionally, SPARQL queries were applied to structure and correlate relevant information, enabling inferences about connections between individuals and events. The results indicate that the combined application of text miningText mining and ontologies enables the identification of behavioral patterns, reconstruction of eventsCriminal event reconstruction, and detection of hidden relationships between interlocutors. The approach was able to recover deleted messages through the merging of backups and to identify critical content, even when it was masked by coded terms or abbreviations. Practical cases demonstrated that the proposed tool offers a low-cost, easy-to-use alternative for non-specialist investigators, while maintaining the technical robustness required for forensic use. The integration of text miningText mining, ontologies, and Semantic Web technologies represents a significant advance for criminal investigations by enabling the structured and contextualized treatment of heterogeneous data. This strategy enhances decision-making, reduces analysis time, and increases the predictive capacity of investigations, contributing to more effective promotion of justice.