Detecting online child grooming conversation
Abstract
Massive proliferation of social media has opened possibilities for perpetrator to conduct the crime of online child grooming. Because the pervasiveness of the problem scale, it may only be tamed effectively and efficiently by using an automatic grooming conversation detection system. Previously, Pranoto, Gunawan, and Soewito [1] had developed a logistic model for the purpose and the model was able to achieve 95% detection accuracy. The current study intends to address the issue by using Support Vector Machine and k-nearest neighbors classifiers. In addition, the study also proposes a low-computational cost classification method on the basis of the number of the existing grooming conversation characteristics. All proposed methods are evaluated using 150 conversation texts of which 105 texts are grooming and 45 texts are non-grooming. We identify that grooming conversations possess 17 features of grooming characteristics. The results suggest that the SVM and k-NN are able to identify grooming conversations at 98.6% and 97.8% of the level of accuracy. Meanwhile, the proposed simple method has 96.8% accuracy. The empirical study also suggests that two among the seventeen characteristics are insignificant for the classification.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Some works have explored various phases of grooming to investigate different themes in online conversations, while others focused on predatory conversation detection, predatory identification, or author profiling.
“Some works [28,33,40,42,43] have explored various phases of grooming to investigate the different themes in online conversations, while others focused on predatory conversation detection [49][50][51], predatory identification [41,52,53], or author profiling [44][45][46][47][48].”
From Online grooming detection: A comprehensive survey of child exploitation in chat logs (2023) · cites this paper as [40]
Once a predator believes that the victim might be emotionally and mentally ready, the conversation goes over to the sexual stage, where sexual topics can be brought up by asking questions such as 'Did you ever touch yourself?'.
“Once a predator believes that the victim might be emotionally and mentally ready, the conversation goes over to the sexual stage, where sexual topics can be brought up by asking questions such as 'Did you ever touch yourself?' [40,42].”
From Online grooming detection: A comprehensive survey of child exploitation in chat logs (2023) · cites this paper as [40]
The exclusive stage is where the predator has assessed the risk, and the theme of the conversation changes.
“The exclusive stage is where the predator has assessed the risk, and the theme of the conversation changes [40,42].”
From Online grooming detection: A comprehensive survey of child exploitation in chat logs (2023) · cites this paper as [40]
Most computer science literature focuses on detecting perpetrators' behaviors using machine learning and natural language processing techniques.
“To combat cybergrooming, researchers have developed numerous computational approaches, with most computer science literature focuses on detecting perpetrators' behaviors using machine learning (ML) and natural language processing (NLP) techniques (e.g., [5, 16-19, 3234, 42, 44, 48, 59, 62, 72, 76-78]).”
From From Vulnerable to Resilient: Examining Parent and Teen Perceptions on How to Respond to Unwanted Cybergrooming Advances (2026) · cites this paper as [42]
The citation lists haven’t been fetched yet
References 16 · Cited by 17
- Authors
- Gunawan, F. E., Ashianti, L., Candra, S., & Soewito, B.
- Affiliations
- Binus University (ID)
- Venue
- 2016 11th International Conference on Knowledge, Information and Creativity Support Systems (KICSS)
- Publisher
- IEEE
- Published
- 2016-11
- Pages
- 1-6
- OpenAlex
- W2678424790
- Cited by
- 17
- References
- 16
- Citations by year
- 2025: 1 · 2024: 1 · 2022: 4 · 2021: 5 · 2019: 5 · 2018: 1
- FWCI
- 1.08
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- Knowledge, Information, and Creativity Support Systems (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Text and Document Classification Technologies
- Physical Sciences › Computer Science › Artificial Intelligence › Hate Speech and Cyberbullying Detection
- Physical Sciences › Computer Science › Artificial Intelligence › Authorship Attribution and Profiling
- Keywords
- online child grooming, child sexual abuse prevention, social media, text classification, support vector machine, k-nearest neighbors, logistic regression
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
