Influence Assessment of Pornography-Related Cryptograms Based on ChatGPT-4o
DOI:
https://doi.org/10.54097/zvhh5r29Keywords:
Network Monitoring Strategy, Pornographic Cryptic Expressions Propagation, Cryptic Expressions Analysis, Influence Assessment, ChatGPT-4o.Abstract
In order to optimise the network monitoring strategy, objectively and accurately analyse the propagation trend and characteristics of different types of pornographic cryptic expressions, and identify the cryptic expressions with greater influence, this article propose a method to assess the influence of pornographic cryptic expressions based on ChatGPT-4o. With the help of knowledge graph to make the content of cryptic text structured, and the frequency of cryptic appearances, the transmission path, and the number of people involved as the factors for assessing the influence of cryptic phrases, the pornography-related cryptic phrases are quantified. Firstly, this article use ChatGPT-4o to preprocess data and extract knowledge from the text corpus of pornography-related cryptic phrases through multiple rounds of Q&A, and construct a knowledge graph in the field of pornography-related cryptic phrases in a low-resource, fast and timely manner, and statistically analyse the frequency of cryptic phrases, propagation paths, the number of people involved and other types of factors based on the knowledge graph, and abstract the impact of different cryptic phrases into impact factors, and use the impact factors to depict the propagation trend and propagation characteristics for comprehensive analysis. The impact of different types of anaphors is abstracted into influence factors, and the influence factors are used to portray the propagation trend and characteristics for comprehensive analysis and judgement. By analyzing the existing cryptic data and calculating the impact factor of cryptic types, we obtained the changes of the impact factor of different cryptic types, verified the scientificity and validity of the impact factor calculation method, and provided a new method and idea for the evaluation of pornographic cryptic language. Combining the advantages of ChatGPT-4o and Knowledge Graph helps to grasp the change trend of cryptic communication in a timely manner, provides powerful support and guidance for network regulation, and is of great significance for protecting social morality and youth's health.
Downloads
References
[1] Ye J, Chen X, Xu N, et al. A Comprehensive Capability Analysis of GPT-3 and GPT-3.5 Series Models [EB/OL]. 2023.
[2] Gilardi F, Alizadeh M, Kubli M. ChatGPT Outperforms Crowd-Workers for Text-Annotation Tasks [EB/OL]. 2023.
[3] Kuzman T, Mozetic I, Ljubesic N. ChatGPT: Beginning of an End of Manual Linguistic Data Annotation? Use Case of Automatic Genre Identification [EB/OL]. 2023.
[4] Huang F, Kwak H, AN J. Is ChatGPT Better than Human Annotators? Potential and Limitations of ChatGPT in Explaining Implicit Hate Speech [EB/OL]. 2023.
[5] Kutela B, Msechu K, Novat N, et al. Uncovering the Influence of ChatGPT's Prompts on Scientific Writings Using Machine Learning-Based Text Mining Approaches [EB/OL]. 2023.
[6] Gao T, Yao X, Chen D. Simcse: Simple Contrastive Learning of Sentence Embeddings [C]//Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, 2021: 6894 - 6910.
[7] Barbieri F, Camacho-Collados J, Anke L E, et al. TweetEval: Unified Benchmark and Comparative Evaluation for Tweet Classification [C]//Findings of the Association for Computational Linguistics: EMNLP 2020, 2020: 1644 - 1650.
[8] Van Hee C, Lefever E, Hoste V. Irony Detection in English Tweets: SemEval-2018 Task 3 [C]//Proceedings of the 12th International Workshop on Semantic Evaluation, 2018: 39 - 50.
[9] Deepa M D. Bidirectional encoder representations from transformers (BERT) language model for sentiment analysis task [J]. Turkish Journal of Computer and Mathematics Education (TURCOMAT), 2021, 12 (7): 1708 - 1721.
[10] Wang Y, Yao Q, Kwok J T, et al. Generalizing from a few examples: A survey on few-shot learning [J]. ACM Computing Surveys (csur), 2020, 53 (3): 1 - 34.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Highlights in Science, Engineering and Technology

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.







