Top 20 Keywords Extracted from Dark Web CSAM Content Using Word2Vec

A list of 20 keywords extracted from child sexual abuse material (CSAM) related content on the dark web. The data was collected from 2,414 dark web pages indexed by the Torch search engine and processed using Word2Vec and other text mining techniques. The dataset was created by Jin Gyeong Kim and last updated on 2026-05-08.

Use Cases

• Benchmarking keyword extraction methods based on the comparative analysis of TF-IDF, Eigenvector Centrality, and Word2Vec described in the study
• Training or evaluating semantic similarity models for investigative keyword expansion based on the Word2Vec methodology
• Developing hybrid keyword ranking models based on the described integration of Eigenvector Centrality with Word2Vec
• Improving the retrieval performance of CSAM-related sites in dark web search experiments based on the evaluated keyword sets

Strengths

• Data is empirically collected from 2,414 dark web pages, providing a domain-specific foundation
• Keywords are the result of a comparative analysis of multiple text mining techniques (TF-IDF, Eigenvector Centrality, Word2Vec)
• The dataset is small (5.5 KB) and focused, containing a curated list of 20 top keywords

Limitations

• Row count is unknown, which may limit suitability assessment
• Column-level documentation is absent; field semantics must be inferred after download
• The dataset's scope is limited to 20 keywords, which may not represent the full semantic field

Provenance

Source

Data collected from dark web pages indexed by the Torch search engine.

Collection Method

Collected using a custom crawler; keywords extracted and compared using TF-IDF, Eigenvector Centrality, and Word2Vec techniques.

Freshness

Last updated 2026-05-08 17:23:29; freshness should be verified