Sorting Out the Bad Seeds: Automatic Classification of Cryptocurrency Abuse Reports

Abuse reporting services collect reports about abuse victims have suffered. Accurate classification of the submitted reports is fundamental to analyzing the prevalence and financial impact of different abuse types (e.g., sextortion, investment, romance). Current classification approaches are problem...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Gomez, Gibran, van Liebergen, Kevin, Sanvito, Davide, Siracusano, Giuseppe, Gonzalez, Roberto, Caballero, Juan
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Gomez, Gibran
van Liebergen, Kevin
Sanvito, Davide
Siracusano, Giuseppe
Gonzalez, Roberto
Caballero, Juan
description Abuse reporting services collect reports about abuse victims have suffered. Accurate classification of the submitted reports is fundamental to analyzing the prevalence and financial impact of different abuse types (e.g., sextortion, investment, romance). Current classification approaches are problematic because they require the reporter to select the abuse type from a list, assuming the reporter has the necessary experience for the classification, which we show is frequently not the case, or require manual classification by analysts, which does not scale. To address these issues, this paper presents a novel approach to classify cryptocurrency abuse reports automatically. We first build a taxonomy of 19 frequently reported abuse types. Given as input the textual description written by the reporter, our classifier leverages a large language model (LLM) to interpret the text and assign it an abuse type in our taxonomy. We collect 290K cryptocurrency abuse reports from two popular reporting services: BitcoinAbuse and BBB's ScamTracker. We build ground truth datasets for 20K of those reports and use them to evaluate three designs for our LLM-based classifier and four LLMs, as well as a supervised ML classifier used as a baseline. Our LLM-based classifier achieves a precision of 0.92, a recall of 0.87, and an F1 score of 0.89, compared to an F1 score of 0.55 for the baseline. We demonstrate our classifier in two applications: providing financial loss statistics for fine-grained abuse types and generating tagged addresses for cryptocurrency analysis platforms.
doi_str_mv 10.48550/arxiv.2410.21041
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2410_21041</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2410_21041</sourcerecordid><originalsourceid>FETCH-arxiv_primary_2410_210413</originalsourceid><addsrcrecordid>eNpjYJA0NNAzsTA1NdBPLKrILNMzMgEKGBkamBhyMgQH5xeVZOalK_iXliiUZKQqOCWmKASnpqYUWyk4lpbk5yaWZCYrOOckFhdnpmUmA3n5eQr5aQrORZUFJfnJpUVFqXnJlQqOSaXFqQpBqQVA04p5GFjTEnOKU3mhNDeDvJtriLOHLtj6-IKizNzEosp4kDPiwc4wJqwCAJtZPXs</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Sorting Out the Bad Seeds: Automatic Classification of Cryptocurrency Abuse Reports</title><source>arXiv.org</source><creator>Gomez, Gibran ; van Liebergen, Kevin ; Sanvito, Davide ; Siracusano, Giuseppe ; Gonzalez, Roberto ; Caballero, Juan</creator><creatorcontrib>Gomez, Gibran ; van Liebergen, Kevin ; Sanvito, Davide ; Siracusano, Giuseppe ; Gonzalez, Roberto ; Caballero, Juan</creatorcontrib><description>Abuse reporting services collect reports about abuse victims have suffered. Accurate classification of the submitted reports is fundamental to analyzing the prevalence and financial impact of different abuse types (e.g., sextortion, investment, romance). Current classification approaches are problematic because they require the reporter to select the abuse type from a list, assuming the reporter has the necessary experience for the classification, which we show is frequently not the case, or require manual classification by analysts, which does not scale. To address these issues, this paper presents a novel approach to classify cryptocurrency abuse reports automatically. We first build a taxonomy of 19 frequently reported abuse types. Given as input the textual description written by the reporter, our classifier leverages a large language model (LLM) to interpret the text and assign it an abuse type in our taxonomy. We collect 290K cryptocurrency abuse reports from two popular reporting services: BitcoinAbuse and BBB's ScamTracker. We build ground truth datasets for 20K of those reports and use them to evaluate three designs for our LLM-based classifier and four LLMs, as well as a supervised ML classifier used as a baseline. Our LLM-based classifier achieves a precision of 0.92, a recall of 0.87, and an F1 score of 0.89, compared to an F1 score of 0.55 for the baseline. We demonstrate our classifier in two applications: providing financial loss statistics for fine-grained abuse types and generating tagged addresses for cryptocurrency analysis platforms.</description><identifier>DOI: 10.48550/arxiv.2410.21041</identifier><language>eng</language><subject>Computer Science - Computation and Language ; Computer Science - Cryptography and Security</subject><creationdate>2024-10</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2410.21041$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2410.21041$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Gomez, Gibran</creatorcontrib><creatorcontrib>van Liebergen, Kevin</creatorcontrib><creatorcontrib>Sanvito, Davide</creatorcontrib><creatorcontrib>Siracusano, Giuseppe</creatorcontrib><creatorcontrib>Gonzalez, Roberto</creatorcontrib><creatorcontrib>Caballero, Juan</creatorcontrib><title>Sorting Out the Bad Seeds: Automatic Classification of Cryptocurrency Abuse Reports</title><description>Abuse reporting services collect reports about abuse victims have suffered. Accurate classification of the submitted reports is fundamental to analyzing the prevalence and financial impact of different abuse types (e.g., sextortion, investment, romance). Current classification approaches are problematic because they require the reporter to select the abuse type from a list, assuming the reporter has the necessary experience for the classification, which we show is frequently not the case, or require manual classification by analysts, which does not scale. To address these issues, this paper presents a novel approach to classify cryptocurrency abuse reports automatically. We first build a taxonomy of 19 frequently reported abuse types. Given as input the textual description written by the reporter, our classifier leverages a large language model (LLM) to interpret the text and assign it an abuse type in our taxonomy. We collect 290K cryptocurrency abuse reports from two popular reporting services: BitcoinAbuse and BBB's ScamTracker. We build ground truth datasets for 20K of those reports and use them to evaluate three designs for our LLM-based classifier and four LLMs, as well as a supervised ML classifier used as a baseline. Our LLM-based classifier achieves a precision of 0.92, a recall of 0.87, and an F1 score of 0.89, compared to an F1 score of 0.55 for the baseline. We demonstrate our classifier in two applications: providing financial loss statistics for fine-grained abuse types and generating tagged addresses for cryptocurrency analysis platforms.</description><subject>Computer Science - Computation and Language</subject><subject>Computer Science - Cryptography and Security</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNpjYJA0NNAzsTA1NdBPLKrILNMzMgEKGBkamBhyMgQH5xeVZOalK_iXliiUZKQqOCWmKASnpqYUWyk4lpbk5yaWZCYrOOckFhdnpmUmA3n5eQr5aQrORZUFJfnJpUVFqXnJlQqOSaXFqQpBqQVA04p5GFjTEnOKU3mhNDeDvJtriLOHLtj6-IKizNzEosp4kDPiwc4wJqwCAJtZPXs</recordid><startdate>20241028</startdate><enddate>20241028</enddate><creator>Gomez, Gibran</creator><creator>van Liebergen, Kevin</creator><creator>Sanvito, Davide</creator><creator>Siracusano, Giuseppe</creator><creator>Gonzalez, Roberto</creator><creator>Caballero, Juan</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20241028</creationdate><title>Sorting Out the Bad Seeds: Automatic Classification of Cryptocurrency Abuse Reports</title><author>Gomez, Gibran ; van Liebergen, Kevin ; Sanvito, Davide ; Siracusano, Giuseppe ; Gonzalez, Roberto ; Caballero, Juan</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-arxiv_primary_2410_210413</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Computer Science - Computation and Language</topic><topic>Computer Science - Cryptography and Security</topic><toplevel>online_resources</toplevel><creatorcontrib>Gomez, Gibran</creatorcontrib><creatorcontrib>van Liebergen, Kevin</creatorcontrib><creatorcontrib>Sanvito, Davide</creatorcontrib><creatorcontrib>Siracusano, Giuseppe</creatorcontrib><creatorcontrib>Gonzalez, Roberto</creatorcontrib><creatorcontrib>Caballero, Juan</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Gomez, Gibran</au><au>van Liebergen, Kevin</au><au>Sanvito, Davide</au><au>Siracusano, Giuseppe</au><au>Gonzalez, Roberto</au><au>Caballero, Juan</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Sorting Out the Bad Seeds: Automatic Classification of Cryptocurrency Abuse Reports</atitle><date>2024-10-28</date><risdate>2024</risdate><abstract>Abuse reporting services collect reports about abuse victims have suffered. Accurate classification of the submitted reports is fundamental to analyzing the prevalence and financial impact of different abuse types (e.g., sextortion, investment, romance). Current classification approaches are problematic because they require the reporter to select the abuse type from a list, assuming the reporter has the necessary experience for the classification, which we show is frequently not the case, or require manual classification by analysts, which does not scale. To address these issues, this paper presents a novel approach to classify cryptocurrency abuse reports automatically. We first build a taxonomy of 19 frequently reported abuse types. Given as input the textual description written by the reporter, our classifier leverages a large language model (LLM) to interpret the text and assign it an abuse type in our taxonomy. We collect 290K cryptocurrency abuse reports from two popular reporting services: BitcoinAbuse and BBB's ScamTracker. We build ground truth datasets for 20K of those reports and use them to evaluate three designs for our LLM-based classifier and four LLMs, as well as a supervised ML classifier used as a baseline. Our LLM-based classifier achieves a precision of 0.92, a recall of 0.87, and an F1 score of 0.89, compared to an F1 score of 0.55 for the baseline. We demonstrate our classifier in two applications: providing financial loss statistics for fine-grained abuse types and generating tagged addresses for cryptocurrency analysis platforms.</abstract><doi>10.48550/arxiv.2410.21041</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2410.21041
ispartof
issn
language eng
recordid cdi_arxiv_primary_2410_21041
source arXiv.org
subjects Computer Science - Computation and Language
Computer Science - Cryptography and Security
title Sorting Out the Bad Seeds: Automatic Classification of Cryptocurrency Abuse Reports
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-24T14%3A15%3A04IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Sorting%20Out%20the%20Bad%20Seeds:%20Automatic%20Classification%20of%20Cryptocurrency%20Abuse%20Reports&rft.au=Gomez,%20Gibran&rft.date=2024-10-28&rft_id=info:doi/10.48550/arxiv.2410.21041&rft_dat=%3Carxiv_GOX%3E2410_21041%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true